DRUGS AND THE PHARMACEUTICAL SCIENCES
`
`VOLUME 119
`
`Dermatological
`ami Transdermal
`
`Furmuiaiiuns
`
`Stratum Corneum Lipid
`
`Interaction between
`solute and vehicle
`
`Increasing escaping
`tendency from vehicle
`into stratum corneum
`
`" p ush“ effect
`
`21:. 1.;
`
`
`
`Increasing affinity.r
`fur
`stratum mmeum
`
`Ksr. T
`
`..
`
`..
`pull effect
`
`edited by
`Kenneth A. Walters
`
`Noven Pharmaceuticals, Inc.
`EX2013
`
`000‘
`
`Mylan Tech., Inc. v. Noven Pharma, Inc.
`|PR2018—00174
`
`
`
`

`

`
`
`
`
`ti
`
`An im
`the mi
`
`impnn
`been n
`men
`and in
`EXIE n51
`
`veinpt
`
`have I
`of lt‘al
`in ma
`the bin
`
`strata;
`volurr
`
`underi
`atives
`
`theml:
`
`ISBN: 0-8247-9889«9
`
`This book is printed on acid-free paper.
`
`Headquarters
`Marcel l)Ckl\'Cl'. Inc.
`270 Madison Avenue. New York. NY 10016
`tel; 212~696—900(J; fax: 212—685—454U
`
`Eastern Hemisphere Distribution
`Marcei Dekker, AG
`Hutgasse 4. Pestfach 812. CH—400i BELSL’L Switzerianti
`tel: 41—61—261784821 fax: 41-61-26|~8896
`
`World Wide Web
`httpflwwwdckkencnm
`
`The publisher ulleni discounts on this book when ordered in bulk quantities. For mute ittf'nrmntinn.
`mile to Special SalesfProfcssional Marketing :1: the headquarters address above.
`
`Copyright © 2002 by Marcel Dekker, inc. All Rights Reserved.
`
`Neither this: hook nor any part may be reproduced or transmitted in any form or by any lllL'ilTlS.
`electronic or mechanical. including pimtuenpying‘ microfiiming, and recording. or by any infor-
`mation storage and retrieval system. without permission in writing From the ptibiishei:
`
`Current printing (last digit}:
`IO
`9
`8
`T
`6
`5
`4
`3
`
`2
`
`l
`
`PRINTED [N THE UNITED STATES OF AMERICA
`
`
`
`

`

`
`
`Contents
`
`
`
`
`
`
`
`
`
`
`a} Dermatological Formulation and 'I"1'ansdcrmal Systems
`Kenneth A. Withers and Keith R. Brain
`
`8. Bioavailabilily and Bioequivulcncc of Dermatological
`Formulations
`Christian .S'Ltrbm' and Adrian F. Davis
`
`9.
`
`Scale—up of Dermatological Dosage Forms: A Case for
`Multivariate Optimization and Product Homogeneity
`Ores! Olejnik and Bruce A. Firestone
`
`IO.
`
`Safety Considerations for Dermal and Transderma] Formulations
`Peter J. Dykes and Anthony D. Patna
`
`l |. Transtlermal Delivery and Cutaneous Reactions
`J’ugdixh Singh and Howard I. Mafbach
`
`Index
`
`319
`
`40 l
`
`499
`
`Co
`
`
`
`

`

`338
`
`Walters and Brain
`
`2. Btopharmaceutical Considerations
`
`A fundamental consideration in the development of transdermal therapeutic systems
`is whether the dermal delivery route can provide the requisite bioavailahility 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 modified. if necessary, by the use of
`penetration enhancers, bttt 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 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 (6]), it is essential, at this stage,
`to ensure that account is taken of the inherent variability in human skin permeation.
`Factorial design and artificial neural networks have been used in the optimi
`zation of transrlerinal drug delivery formulations using in vitro skin permeation tech-
`niques (62-64). For example, Kartdimalla 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 pres
`liminary formulation optimization.
`A major drawback of tt‘ansdermal delivery systems is the potential for localized
`irritant and allergic cutaneous reactions. At the earlier stages of formulation devel-
`opment, it is. therefore,
`important to evaluate both drugs and cxcipients for their
`potential to cause irritation and sensitization (sec Chapters l0 and 11). This is true
`for all transdermal systems, but especially for those that may stay in place for pro~
`longed periods. The degree of primary and chronic irritation, and the potential
`to
`cause contact allergy, photoirritation. and pltotrmllergy should be determined. Norm
`malty. the drug and exeipicnts 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
`
`7——
`
`Drug Fc
`
`Table 6
`
`Apalicat
`vehicle“
`
`
`
`W:_-.'_P [2
`W5; (40
`W1P l-‘lU
`
`"Flux wzt
`methods
`“W, watct
`'Data are
`50mm; Ii
`
`always
`Fortuna
`and mil
`
`3. De
`
`All palt
`three b;
`and dl‘U
`
`by a rat
`of trans
`
`ing the
`which :
`
`designe
`from a
`adhesiv
`mation
`mers, t
`Additio
`interfer
`other ft
`and do:
`
`system
`selectic
`ion] or
`A
`adhesiv
`can he
`
`rheolog
`sive to
`stretch
`formati
`and Ion
`initial 2
`residue
`
`
`
`

`

`Drug Formulation and Transdermal Systems
`
`339
`
`
`
`Table 6 Experimental Versus Predicted Flux" of Melatonin
`
`Application
`Experimental
`ANN Prediction
`RSM Prediction
`
`vehicle”
`t‘ttgz'crn2 h l)"
`(lug/’cm'i h J)
`[,ttgfcrn2 h ')
`
`l2.73
`[2.11r
`”.32 i 0.86
`W1E2Pf'2026f122fl)
`11.76
`11.83
`10.89 i 1.36
`W:Ii (40:60}
`
`
`
`1541 l 1.39 6,75WEP (4015(1) 150
`
`"Flux was predicted on the basis of artificial neural networks {ANN} or response surface
`methods (RSM).
`"W, water: E. ethanol; 1’. propylene glycol.
`:Data are means _- standard deviation (:1 = 3} and represent flux through rat dorsal skin.
`Source: Ref. («'13.
`
`always be carried out using the final and complete fm'nmlalion in human volunteers.
`Fortunately, most of the observed skin reactions to transdermal systems are transient
`and miltl 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 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 nosins 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. (3d). There are three critical consideratimis 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. aci'ylatcs, 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 adhe—
`sive to flow into 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 enough to 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
`
`
`
`

`

`34a
`
`(3) DRUG [N ADHESIVE
`
`Walters and Brain
`
`Adhesive Layer
`
`Backing Layer
`
`Release Liner
`
`(b) DRUG IN MATRIX
`
`(C) DRUG IN RESERVOIR
`
`(d) PERIPHERAL ADHESIVE
`
`,
`Backing Layer
`Polymeric Matrix
`Adhesive Layer
`Release Liner
`
`Backin L
`g ayer
`Reservoir
`
`Membrane
`
`Adhesive Layer
`
`Release Liner
`
`Backing Layer
`
`Adhesive Layer
`.
`Reservoir
`Membrane
`eease In r
`l
`L'
`e
`
`F!
`
`Figure 6
`
`'l'ypical 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 oeclusivity and round
`patches tend to be more secure than those ol’ sharply angled geometry. If the patch
`is able to conform to the skin contours it resists lilting and buckling with movement.
`The presence of water may affect adhesive properties; therefore. the oeelusivity of
`the system must be taken into consideration. Occlusion for prolonged periods can
`
`Drug F0
`
`lead to e
`
`increaseTh
`material:
`acetate.
`to the dr
`those th;
`
`water lo
`subpatcr
`layer wi'
`films or
`rubber—b
`adhesive
`
`4. Dru
`
`The thre
`
`system I
`in—adhes
`for exam
`combine
`
`M‘
`drug res:
`size ant
`
`through
`material:
`
`eopotyn‘.
`
`aeryloni'
`haneer (
`also bee
`
`drug (6.6
`brane,
`is
`mulated
`A“
`.
`dermal
`Fabi‘icatc
`nitroglyt
`ture and
`hexane].
`amount
`line—turd
`the pore
`to
`membra
`
`evaluate
`membra
`widely t
`eluded tl
`the t‘abr
`
`
`
`

`

` Drug Formulation and Transdermal Systems
`
`
`341
`
`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 polyvinylchloridc, 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 sufficient resistance to trausepidermal
`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 arty of the adhesive to which it is bound. Liners are usually
`films or coated papers, and silicone. release coatings are used with acrylate» and
`rubber—based adhesive systems. whereas fluorocarbon coatings are used with silicone
`adhesives (65).
`
`4. Drug and Enhancer Incorporation
`
`The three principal methods for incorporating the. active species into a transdcrmal
`system have led to the loose classification of patches as membrane, matrix, or drug—
`in—adhcsivc 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 membrane—matrix device to deliver an initial bolus dose.
`Membrane patches contain a delivery rate—controlling nrembrane between the
`drug reservoir and the skin. Microporous membranes, which control drug flux by the
`size and tortrtosity of pores in the membrane. or dense polymeric membranes,
`through which the drug permeates by dissolution and diffusion, may be used. Several
`materials can he used as rate—controlling membranes (e.g.. ethylene-vinyl acetate
`copolynrers. silieorres, high—density polyethylene, polyester elastotners, and poly—
`acrylonitrile). Ideally, the membrane should be permeable only to the drug and en—
`hancer (if present) and should retain other formulation excipicnts. Membranes have
`also been designed such that they allow differential permeation of an enhancer and
`drug (66- ()8). 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—methyl-l—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 (butanel) used, together with the solvent (cyclo-
`hcxarre).
`in the casting process. This is. perhaps, not surprising, as increasing the
`amount of nonsolvent increases the porosity of the east membrane. The concept of
`tine—tuning delivery of a drug through a given membrane by subtle adjustment of
`the porosity creates some exciting new possibilities in transderrttal technology (70).
`In all marketed membrane—controlled transdcrmal systems, the rate—controlling
`membrane is fabricated from synthetic polymeric materials. Thaeharodi and Rao (71)
`evaluated the potential of two biopolymers (fetal calf skin collagen and clritosan) in
`membrane systems for delivery of nifedipinc. Chitosan tdcaectylated chitin) is a
`widely distributed major constituent of the shells of marine shellfish.
`it was con-
`cluded that the permeability of both biopolymers could be readily adjusted by altering
`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 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 aqueousmal-
`coholic solutions and gels, with or without various cosolvents. and polymeric nia—
`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 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. poly—
`urethanes, polyvinyl alcohol, polyvinylpyn‘olidottcs. 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 difficult, 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 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 of these are illustrated by a study on release of
`several drugs from silicone matrices (73]. Silicone medical-grade elastomers (poly-
`dintethylsiloxanes) are flexible, lipophilic polymers, with excellent compatibility with
`biological
`tissues,
`that can be eoformulated with hydrophilic cxcipients. such as
`glycerol, and inert hllers, such as titanium dioxide. to alter rcicase kinetics. Increasing
`the amount of glycerol
`in the matrix increased the release rate of indomethaein,
`propranelol,
`testosterone. and progesterone, whereas incorporation of inert
`tillers
`(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 polydimethylsiloxalic—polyethylene oxide graft co—
`polymers {74). These data demonstrate that release rates can he modulated to achieve
`a desired profile by simple formulation modification.
`Perhaps the simplest form of transdermai drug. delivery device, which is now
`most commonly employed, is the drug-in-adltesivc 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 film, to produce an adhesive tape,
`This simplicity is, however, deceptive and several factors. involving potential inter-
`action 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 physicoehetnical 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 hydrophilie
`drugs within the adhesive matrix.
`
`Drug F
`
`l1
`
`drug-in
`exalnpl
`in loss
`(75]. A
`was to
`and the
`
`a drug-
`is that
`the ski
`
`drug—in
`resttltin
`
`strengtl
`this lTIE.
`be over
`
`System
`of the
`are ave
`
`serious
`
`ester t)
`lished.
`
`skin [3(-
`copolyt
`Vi
`
`apprcci
`is rele:
`enhanc
`
`may ht
`enhanc
`to obsc
`
`5.
`
`By
`
`The ma
`system:
`the dru
`
`quired)
`Inoldin
`layer,
`1'
`aging. .
`cess: te
`laminat
`tent (7i
`A
`content
`
`mity ci
`assayin
`ing drtt
`Manuf;
`
`
`
`

`

`Drug Formulation and Transderrnal Systems
`
`343
`
`into a
`Incorporation of other excipients. such as skin permeation enhancers.
`tlrug-in-adhesivc system may alter drug, release rates and adhesive properties. For
`example. addition of i% urea to a polyacrylate pressure—sensitive adhesive resulted
`in loss of adl'icsion and skin contact could ttot be maintained over the required period
`(75). A strategy to reduce the influence 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 drug—enhanccr—rclcasing layer. One disadvantage of this type of system, however.
`is that the dmg—cnhancer—releasing layer may not remain in intimate contact with
`the skin. If high levels of liquid skin penetration enhancers are incorpta'at'ed into
`drug-in—adhcsivc transdermal patches,
`there is likely to he 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 separatioa
`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 retnains to be estab—
`lished.
`Il may also be possible to maintain adhesive properties in the presence of
`skin penetration enhancers by using different molecular weight blends of acrylic
`coptiilyiuers (77,78).
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`As with all controlledri‘clease delivery systems, final product checks include
`
`content unift‘irtnity. release-rate determination, and physical testing. Content—unifor-
`
`mity et-‘alnatinn involves removing a random sample of patches from a batch and
`
`assaying the att'tount of drug present. Of the several methods available for determin—
`ing drug release rates from controlled-release formulations, the U.S. Pharmaceutical
`
`Manufacturers Association (Pb/IA) Committee (80) has recommended three:
`the
`
`
`
`is important to
`When enhancers are incorporated into transdermal systems it
`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 influence on local
`toxicity. Much remains to be done in the field 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 Iransdetmal
`systems are.
`to a large extent. similar. All involve the following stages: preparing
`the drug; mixing the drug (with other excipicnts 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 te.g., backing
`layer, rate—controlling membrane, and release liner); die—cutting: and finally. pack-
`aging. Casting and lamination are the most critical 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).
`
`