International Journal of Pharmaceutics 375 (2009) 22–27
`
`Contents lists available at ScienceDirect
`
`International Journal of Pharmaceutics
`
`j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / i j p h a r m
`
`Alteration of the diffusional barrier property of the nail leads to greater
`terbinafine drug loading and permeation
`Anroop B. Nair a, Srinivasa M. Sammeta a, Hyun D. Kim b, Bireswar Chakraborty b,
`Phillip M. Friden b, S. Narasimha Murthy a,∗
`
`a Department of Pharmaceutics, The University of Mississippi, University, MS 38677, United States
`b Transport Pharmaceuticals Inc., Framingham, MA 01701, United States
`
`a r t i c l e
`
`i n f o
`
`a b s t r a c t
`
`Article history:
`Received 26 January 2009
`Received in revised form 4 March 2009
`Accepted 11 March 2009
`Available online 24 March 2009
`
`Keywords:
`Abrasion
`Enhancers
`Iontophoresis
`Nail barrier
`Pretreatment
`Terbinafine
`
`The diffusional barrier property of biological systems varies with ultrastructural organization of the tis-
`sues and/or cells, and often plays an important role in drug delivery. The nail plate is a thick, hard and
`impermeable membrane which makes topical nail drug delivery challenging. The current study inves-
`tigated the effect of physical and chemical alteration of the nail on the trans-ungual drug delivery of
`terbinafine hydrochloride (TH) under both passive and iontophoretic conditions. Physical alterations
`were carried out by dorsal or ventral nail layer abrasion, while chemical alterations were performed by
`defatting or keratolysis or ionto-keratolysis of the nails. Terbinafine permeation into and across the nail
`plate following various nail treatments showed similar trends in both passive and iontophoretic delivery,
`although the extent of drug delivery varied with treatment. Application of iontophoresis to the abraded
`nails significantly improved (P < 0.05) TH permeation and loading compared to abraded nails without ion-
`tophoresis or normal nails with iontophoresis. Drug permeation was not enhanced when the nail plate
`was defatted. Keratolysis moderately enhanced the permeation but not the drug load. Ionto-keratolysis
`enhanced TH permeation and drug load significantly (P < 0.05) during passive and iontophoretic delivery
`as compared to untreated nails. Ionto-keratolysis may be more efficient in permeabilization of nail plates
`than long term exposure to keratolysing agents.
`
`© 2009 Elsevier B.V. All rights reserved.
`
`1. Introduction
`
`Ungual and trans-ungual drug delivery continues to receive sig-
`nificant attention due to the need for efficacious topical therapies
`for onychomycosis given the potential risk of systemic adverse
`effects associated with the conventional oral therapy (Effendy,
`1995). The major concern in topical therapy is the low trans-nail
`penetration into the deeper nail stratums because of the inherent
`limitation of low permeability of the keratinised nail plates (Baran
`and Kaoukhov, 2005). In one approach to address this issue, per-
`meation enhancers were screened for their ability to enhance the
`trans-ungual permeation of antifungal agents (Van Hoogdalem et
`al., 1997; Kobayashi et al., 1998; Malhotra and Zatz, 2002; Hui et al.,
`2003; Hao et al., 2008). In another approach, drug discovery groups
`have synthesized and screened newer antifungal agents with low
`keratin binding and higher nail permeation properties (Tatsumi et
`al., 2002; Hui et al., 2007). Despite all these efforts, the success rates
`
`∗ Corresponding author at: 113, Faser Hall, School of Pharmacy, Department of
`Pharmaceutics, The University of Mississippi, University, MS 38677, United States.
`Tel.: +1 662 915 5164; fax: +1 662 915 1177.
`E-mail address: murthy@olemiss.edu (S.N. Murthy).
`
`0378-5173/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
`doi:10.1016/j.ijpharm.2009.03.012
`
`of topical therapies have so far been disappointing in comparison
`to systemic therapy (Murdan, 2008).
`The barrier properties of a particular tissue are dictated by char-
`acteristics such as membrane composition and thickness, and the
`specific routes available for drug permeation through the tissue. In
`comparing nail and skin, for example, nail is a thick, hard and com-
`pact keratin membrane with low lipid content (≤1%, w/w) whereas
`stratum corneum, which provides the primary barrier for skin, is
`thin (10–20 ␮m), highly flexible and has a high lipid content (10%)
`(Gupchup and Zatz, 1999). Additionally, the pathways available for
`drug permeation across the skin and nail differ; for example, the fol-
`licular route in skin is absent in nail (De Berker et al., 2007). Taken
`together, these characteristics of the nail make it a formidable bar-
`rier to drug permeation and the challenge to improve topical deliv-
`ery of drugs into and through the nail remains formidable as well.
`Recently the ability of iontophoresis to enhance the trans-
`ungual permeation of salicylic acid and terbinafine was demon-
`strated (Murthy et al., 2007a; Nair et al., 2009). In addition, the
`perm-selective nature of the human nail plate was found to be com-
`parable to that of skin (Murthy et al., 2007b). Iontophoresis uses a
`low level electric current to actively facilitate drug transport across
`biological membranes (Batheja et al., 2006). The enhanced perme-
`ation of charged drug molecules by iontophoresis was principally
`
` Page 1
`
`Anacor Exhibit 2004
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00169
`
`

`

`A.B. Nair et al. / International Journal of Pharmaceutics 375 (2009) 22–27
`
`23
`
`due to electrorepulsion with a lesser contribution by electroos-
`mosis (Hao and Li, 2008a). The influence of variables such as
`current level and density, vehicle pH, type of electrode, co-ions,
`drug concentration, and charge on the transport of drugs dur-
`ing iontophoresis in transdermal delivery is well-known (Kalia et
`al., 2004). A similar influence/effect of the above said variables
`in trans-ungual iontophoretic delivery was also assessed in ear-
`lier studies. It was observed that the increase in current density
`(0.1–1 mA/cm2) enhanced the drug permeation linearly (Murthy et
`al., 2007a; Nair et al., in press). The role of drug concentration, ionic
`strength and vehicle pH on trans-ungual iontophoretic permeation
`was also investigated (Murthy et al., 2007a). Furthermore, Hao and
`Li (2008b) reported that pH and ionic strength had little effect on
`electroosmotic transport in trans-ungual iontophoresis.
`Terbinafine is a potent antifungal agent which belongs to the
`allylamine class of antifungals, is highly effective in treating der-
`matophyte infections, and is the current treatment of choice
`in onychomycosis (Darkes et al., 2003). It possesses low min-
`imum inhibitory concentrations (∼0.001–0.01 ␮g/mL) and low
`minimal fungicidal concentrations (∼0.003–0.006 ␮g/mL) against
`dermatophytes (Darkes et al., 2003). Terbinafine acts by block-
`ing ergosterol biosynthesis by inhibiting the enzyme squalene
`epoxidase (fungistatic), which also leads to the toxic accumula-
`tion of intra-cellular squalene thereby exhibiting fungicidal activity
`(Ryder, 1992; Darkes et al., 2003). In the present study, the influence
`of barrier alteration on the passive and iontophoretic trans-ungual
`delivery (both permeation and drug load) was systematically inves-
`tigated using terbinafine as a model drug.
`
`2. Materials and methods
`
`2.1. Materials
`
`Terbinafine hydrochloride (TH) [MW = 327.90 Da, aqueous solu-
`bility = 1 mg/mL, pKa = 7.1, log octanol/water partition coefficient of
`terbinafine (Alberti et al., 2001) = 3.3], was procured from Uquifa,
`Jiutepac, Mexico. Sodium sulfite and salicylic acid were purchased
`from Sigma–Aldrich, St. Louis, MO. Human cadaver nails, both male
`and female, aged between 49 and 86 years, with varying thickness
`of 0.4–0.7 mm were procured from Science care (Phoenix, AZ) and
`were stored at 4 ◦C until used. All other chemicals and reagents used
`were of analytical grade. All solutions were prepared in deionized
`water.
`
`2.2. Analytical method
`
`The amount of terbinafine in the samples was quantified by high
`performance liquid chromatography (HPLC) system (Waters, 1525)
`with an autosampler (Waters, 717 plus) consisting of a Phenomenex
`C18 (2) 100 R analytical column (4.6 mm× 150 mm, Luna, 5.0 ␮m)
`and a variable wavelength dual (cid:2) absorbance detector (Waters,
`2487). Mobile phase consisted of aqueous solution (0.096 M tri-
`ethyl amine, 0.183 M orthophosphoric acid) and acetonitrile (60:40)
`adjusted to pH 2 with orthophosphoric acid. Elution was performed
`isocratically at 32 ◦C at a flow rate of 1.0 mL/min. Injection volume
`was 20 ␮L and the column effluent was monitored at 224 nm. The
`method was validated by determination of linearity, precision, and
`accuracy. The range for the calibration curve was 2–1000 ng/mL
`(R2 = 0.99). The coefficient of variation and the accuracy ranged
`1.03–6.08% and −0.54 to −6.96%, respectively.
`
`2.3. Nail treatments
`
`2.3.1. Abrasion of dorsal or ventral nail layer
`Nails were cleaned and adherent tissues were removed. The
`dorsal or ventral surface of the nail was physically abraded using
`
`sandpaper (grade #180) until the top or bottom layer (3/10 of the
`total thickness) was removed. The complete removal of the dor-
`sal/ventral layer was confirmed by microscopic examination of
`sections of the nail plate.
`
`2.3.2. Defatting and keratolysing of nails
`Cleaned nail pieces were defatted by placing them in a beaker
`containing chloroform:methanol (2:1) mixture (10 mL) and stirred
`for a period of 12 h. Similarly, nail pieces were stirred in salicylic
`acid solution (4 mg/mL) for 12 h to keratolyse the nails. Treatment
`period (12 h) was selected since the longer exposure (24 h) of nails
`with salicylic acid solution did not improve the permeation.
`
`2.3.3. Pretreatment with keratolytic agents in conjunction with
`iontophoresis (ionto-keratolysis)
`Each of the nail plates were soaked in 0.9% (w/v) saline for 1 h,
`cleaned and mounted on a nail adapter (PermeGear, Bethlehem,
`PA). The whole assembly was sandwiched between the two cham-
`bers in a Franz diffusion cell (Logan Instruments Ltd., Somerset, NJ).
`Each 500 ␮L solution of salicylic acid (4 mg/mL) or sodium sulfite
`(50 mg/mL) was placed in the donor compartment, and 5 mL of nor-
`mal saline in the receiver. Cathodal iontophoresis was carried out by
`applying a constant current of 0.5 mA/cm2 by placing the cathode
`(silver chloride electrode) in the donor compartment and the anode
`(silver electrode) in the receiver, for a period of 1 h. The nail plate
`was rinsed with pH 3 water before carrying out the permeation
`experiments.
`
`2.4. Permeation studies with terbinafine
`
`Nail plates were soaked in 0.9% (w/v) saline for 1 h immediately
`prior to use and mounted on a nail adapter (PermeGear, Bethle-
`hem, PA). The whole assembly was sandwiched between the two
`chambers in a Franz diffusion cell (Logan Instruments Ltd., Som-
`erset, NJ). TH solution (500 ␮L, 1 mg/mL adjusted to pH 3 using
`0.01N HCl) was placed in the donor compartment. The active dif-
`fusion area exposed to both the donor and receiver compartments
`was 0.2 cm2. The receptor compartment, which had a capacity of
`5 mL and was filled with saline (adjusted to pH 3 using 0.01N HCl)
`which provides sink conditions due to increased drug solubility. The
`receiver compartment was stirred at 600 rpm with a 3-mm mag-
`netic stir bar at room temperature. Samples were withdrawn from
`the receiver compartment after 24 h and analyzed for terbinafine
`concentration and the cumulative amount of terbinafine permeated
`into the receiver chamber normalized to the surface area exposed
`to the drug was expressed as ␮g/cm2.
`Anodal iontophoresis was carried out by fixing 0.5 mm diameter
`Ag/AgCl wire electrodes (Alfa Aesar, Wardhill, MA) at a distance of
`2 mm from the nail surface in donor and receiver chambers. Iomed
`Phoresor II dose controller (Iomed Inc., Salt Lake City, UT) was used
`for application of a constant DC. The anode was connected to the
`donor and the cathode to the receiver chamber and a constant cur-
`rent (0.5 mA/cm2) was applied for a period of 24 h.
`
`2.5. Amount of drug in nail
`
`After the in vitro diffusion studies, the nail plates were marked
`for active diffusion area (using permanent marker and metric
`punch), washed with water and alcohol five times each using a
`standardized protocol to avoid the washout of drug loaded in the
`nail while removing surface drug. In brief, washing was carried out
`by holding the nail with forceps and shaking twice by placing in
`2 mL of water (pH 3). Five such washings were performed in fresh
`2 mL of pH 3 water each time. The nail surface was cleaned using
`a cotton swab soaked in 95% ethanol and rinsed with 1 mL ethanol
`(95%)—this alcohol washing procedure was repeated 5 times. The
`
` Page 2
`
`Anacor Exhibit 2004
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00169
`
`

`

`24
`
`A.B. Nair et al. / International Journal of Pharmaceutics 375 (2009) 22–27
`
`Fig. 1. Comparison of the amount of terbinafine permeated at the end of 24 h with various nail treatments. Ionto-keratolysis was carried out by cathodal iontophoresis
`(0.5 mA/cm2) using salicylic acid/sodium sulfite for a period of 1 h. The diffusion area was 0.2 cm2. Data expressed as means± S.D. (n = 4).
`
`nail surface was wiped with Kim wipe, active diffusion area was
`excised, weighed and dissolved in 1 M sodium hydroxide (1.5 mL)
`by constant overnight stirring. Extraction of drug was carried out
`by a slight modification of the method described by Dykes et al.
`(1990). Briefly, after dissolving the nails in the vials, 200 ␮L of 5 M
`hydrochloric acid was added to neutralize the mixture. Terbinafine
`was extracted by adding hexane (3 mL) to the vial and shaking
`manually for 30 min. The mixtures were transferred into centrifuge
`tubes and centrifuged at 4000 rpm for 10 min. The hexane layer
`was collected, 1 mL of 0.5 M sulfuric acid–isopropyl alcohol mix-
`ture (85:15) was added and the mixture was shaken vigorously for
`30 min. The lower acidic aqueous layer, which holds the majority
`of terbinafine, was collected separately and the amount of drug in
`the nail was determined by HPLC. This extraction procedure was
`validated by spiking different drug concentrations (2–20 ␮g/mL)
`into sodium hydroxide solution in which the nail was previously
`dissolved. The recovery was found to be 84± 7%.
`The amount of terbinafine diffused into the lateral region was
`determined by dissecting the lateral nail area (4–5 mm surrounding
`the active diffusion area), washed, dried and weighed. The amount
`of drug was determined as described before.
`
`2.6. Data analysis
`
`Enhancement was calculated as the ratio of the amount of drug
`permeated (␮g/cm2) in 24 h with various treatments to its corre-
`sponding control. Statistical analysis was performed by one-way
`analysis of variance (ANOVA) and t-test using Graphpad prism 5,
`graphpad software, Inc., CA, USA, to test the effects of various treat-
`ments. P value less than 0.05 was considered statistically significant.
`The data points provided in the graphs are an average of four trials.
`The error bars represent the standard deviation.
`
`3. Results and discussion
`
`3.1. Permeation of TH across the nail plate
`
`3.1.1. Passive permeation
`The reports in the literature and our previous study demon-
`strated that the maximum amount of drug is retained in the dorsal
`layer of the nail during permeation across the nail plate (Hui et
`al., 2002; Nair et al., in press) when compared to the intermedi-
`ate and ventral layers. This suggests that the upper most layer of
`the nail plate, which is relatively more dense and harder than the
`other 2 layers (Walters and Flynn, 1983), may reduce the rate of
`
`permeation of drugs into the deeper layers of the nail plate and
`the nail bed. Therefore, removal of the upper layer may improve
`drug delivery across the nail plate. To test this hypothesis, a pas-
`sive permeation study using terbinafine was initially carried out
`for a period of 24 h after the complete removal of the dorsal nail
`layer by abrasion with sandpaper. This process eliminated almost
`3/10th of the total nail thickness. The donor vehicle pH was adjusted
`to 3, as the drug possesses a higher solubility at acidic pH lev-
`els. The amount of drug permeated passively within 24 h across
`the nail plate devoid of the upper layer (abraded nails) during
`delivery was found to be 6.87± 1.26 ␮g/cm2, which was ∼4-fold
`higher than the drug permeation through the intact nail (control:
`1.89± 0.66 ␮g/cm2) (P = 0.0004). In another set of experiments, the
`ventral layer of nail was abraded (3/10th of thickness) and the TH
`permeation was assessed. The results indicated that TH permeation
`was not altered significantly (2.48± 0.91 ␮g/cm2 after passive per-
`meation, P = 0.3343) compared to intact nails without a decrease in
`the nail thickness, suggesting that reduction in nail thickness alone
`did not result in the observed permeation enhancement with the
`dorsal layer abraded nails. These results further confirm that the
`dorsal layer is a significant barrier to drug permeation in topical
`nail drug delivery, which is in agreement with the earlier report
`(Kobayashi et al., 1999). Further, this suggests that abrasion of the
`dorsal nail layer could be used to enhance drug delivery of anti-
`fungal agents during topical therapy. This technique is generally
`reported to be an effective and inexpensive method for treating
`hyperkeratosis of the nail plate or partial removal of the plate to
`resolve haematomas (Di Chiacchio et al., 2003). Although this tech-
`nique possesses many advantages, the application of this in practice
`could be limited to severely diseased nails (Baran et al., 2008).
`Chemical permeation enhancers that are presumed to act on
`the nail plate composition were next examined for their ability to
`increase the permeation of terbinafine across the human nail plate.
`Similar to skin, the nail plate also could have two potential path-
`ways of drug permeation, one being permeation across the keratin
`matrix and the second being permeation through the lipid domain
`which constitutes approximately 1% (w/w) of the nail. TH perme-
`ation studies were carried out across deplipidized or keratolysed
`nail plates. The nail was defatted by soaking the nail plates in a
`chloroform–methanol mixture (2:1) overnight. Similarly, the nail
`plates were kept immersed in salicylic acid solution (4 mg/mL) for
`effective keratolysis. Inconsistent reports exist on the ability of sali-
`cylic acid to enhance the permeation of antifungal agents in topical
`nail delivery (Kobayashi et al., 1998; Murdan, 2002; Malhotra and
`Zatz, 2002). However, in the current study, the amount of TH which
`
` Page 3
`
`Anacor Exhibit 2004
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00169
`
`

`

`A.B. Nair et al. / International Journal of Pharmaceutics 375 (2009) 22–27
`
`25
`
`Fig. 2. The amount of TH loaded (␮g/mg) at 24 h in the active diffusion area of nail during permeation studies with various nail treatments. Ionto-keratolysis was carried out
`by cathodal iontophoresis (0.5 mA/cm2) using salicylic acid/sodium sulfite for a period of 1 h. The diffusion area was 0.2 cm2. Data expressed as mean± S.D. (n = 4).
`
`permeated over 24 h of passive delivery following keratolysis was
`found to be 3.47± 0.52 ␮g/cm2, which is a moderate enhancement
`(∼2-fold, P = 0.0094) over passive delivery in normal nails (Fig. 1).
`It has been proposed that the enhancement in drug permeation by
`salicylic acid pretreatment is likely due to an increase in the nail
`hydration level and the disruption of the keratin tertiary structure
`(Kobayashi et al., 1998). In contrast, the passive drug permeation in
`defatted nails (2.43± 0.94 ␮g/cm2) was not statistically significant
`when compared with untreated nails (P = 0.3834). These results
`suggest that permeation enhancers which disrupt keratin struc-
`ture are likely to improve the delivery of terbinafine. However, the
`agents that interact with the lipids in the nail plate may not be effec-
`tive in enhancing the trans-ungual delivery of TH. It appears that
`the nail plate may not have a continuous lipid pathway running
`through the nail plate which could again be due to the relatively
`small fraction of lipid constituting the nail plate.
`Iontophoresis is a technique in which low level electric cur-
`rent is used to enhance drug delivery across biological membranes.
`Application of a driving force with the same polarity as that of the
`compound to be delivered should enhance its delivery across the
`hydrated nail plate. Recently our group demonstrated the ability
`of short term iontophoresis to enhance the trans-ungual delivery
`of terbinafine (Nair et al., 2009; Nair et al., in press). We later
`hypothesized that the combination of iontophoresis with chem-
`ical permeation enhancers (ionto-keratolysis) could improve the
`effect of the enhancers by driving more of the enhancer into the nail
`matrix. Salicylic acid and sodium sulfite (keratolytic agents) were
`selected and combined with iontophoresis as a pretreatment in the
`permeation studies. These enhancers are anionic in nature and can
`be driven into the nails by the application of cathodal iontophoresis
`(0.5 mA/cm2). Salicylic acid solution (4 mg/mL: pH 5 and pKa 3.1)
`or sodium sulfite solution (50 mg/mL; pH 8.6) was placed (500 ␮L)
`in the donor chamber and cathodal iontophoresis was carried out
`for 1 h as a pretreatment of the nails prior to permeation stud-
`ies with TH. After the ionto-keratolysis, the donor compartment
`enhancer solution was replaced with TH solution (1 mg/mL) and
`the drug was allowed to permeate passively. Ionto-keratolysis for a
`short duration (1 h) with salicylic acid or sodium sulfite enhanced
`the amount of drug permeated within 24 h [5.01± 1.09 ␮g/cm2
`(P = 0.0157) and 4.87± 1.75 ␮g/cm2 (P = 0.0034), respectively] ver-
`sus the normal untreated control nails (1.89± 0.66 ␮g/cm2) (Fig. 1).
`Sodium sulfite is postulated to act by reducing the disulfide link-
`ages in keratin (Malhotra and Zatz, 2002). However, the amount
`of terbinafine permeated in this set of studies was less than the
`amount of drug permeated through the nail plate abraded to
`
`remove the dorsal layer. Interestingly, the amount of passive TH
`permeation after ionto-keratolysis (0.5 mA/cm2 for 1 h) with sali-
`cylic acid was ∼1.5-fold higher (P = 0.0435) than the nails soaked in
`a solution of salicylic acid for 12 h. This shows that ionto-keratolysis
`results in a higher structural alteration as compared to long expo-
`sure of nails to keratolytic agents, which may not be a clinically
`feasible technique.
`
`3.1.2. Iontophoretic permeation
`To investigate the effect of iontophoresis on the permeation of
`terbinafine through the barrier modified nails, anodal iontophore-
`sis (0.5 mA/cm2 for 24 h) was carried out using abraded, defatted,
`keratolysed and ionto-keratolysed (using salicylic acid and sodium
`sulfite for 1 h) nails. The drug solution was delivered at pH 3, which
`also ensured that the drug was fully ionized for optimal delivery
`via electrorepulsion. The amount of terbinafine permeated in 24 h
`through nails subjected to different treatment processes and con-
`trol are depicted in Fig. 1. The significant increase in permeation
`due to iontophoresis was obvious in all the cases when compared
`to passive delivery. Surprisingly, the fold enhancement due to ion-
`tophoresis was comparable (∼5–6-fold) in all the cases, irrespective
`of the nail treatment. The plausible explanation for this constant
`enhancement in drug permeation by iontophoresis could be due
`to the following reasons. In the present study, the experimen-
`tal conditions of iontophoresis (duration, electrical dose and drug
`concentration) were the same in all cases, irrespective of the nail
`treatments. There were no other counterions present in the donor
`drug solution other than H+ ions. Due to the identical experimental
`condition, the ratio of terbinafine ion to H+ ions remains the same
`in all cases. Hence the percentage of total charge transported by
`terbinafine ions remains approximately constant.
`Iontophoretic drug permeation through treated nails showed
`a pattern similar to that observed in passive permeation stud-
`ies (Figs. 1 and 2). As expected, TH permeation was enhanced
`(∼3-fold) in nails without the dorsal layer (37.21± 4.93 ␮g/cm2)
`when compared to the normal untreated nails during iontophoresis
`(P < 0.0001). However, the permeation was not enhanced when the
`ventral nail was abraded (13.06± 3.47 ␮g/cm2, P = 0.3668). Defat-
`ted and keratolysed nails did not significantly enhance terbinafine
`permeation even in conjunction with iontophoretic delivery when
`compared to normal nails. The amount of terbinafine perme-
`ated across the defatted and keratolysed nail was 12.32± 3.12
`(P = 0.5238) and 16.59± 4.73 ␮g/cm2 (P = 0.0812), respectively,
`and was comparable to that obtained with untreated nails
`(10.95± 2.58 ␮g/cm2). Our results suggest that these two passive
`
` Page 4
`
`Anacor Exhibit 2004
`Flatwing Pharmaceuticals, Inc. v. Anacor Pharmaceuticals, Inc
`IPR2018-00169
`
`

`

`26
`
`A.B. Nair et al. / International Journal of Pharmaceutics 375 (2009) 22–27
`
`pretreatment methods could not sufficiently alter the nail barrier
`properties to enhance iontophoretic drug permeation. In contrast,
`ionto-keratolysis for 1 h with salicylic acid or sodium sulfite fol-
`lowed by anodal iontophoresis (24 h) of TH significantly enhanced
`the terbinafine permeation to a similar extent [26.72± 5.16
`(P = 0.0016) and 29.74± 3.61 ␮g/cm2 (P = 0.0001) using salicylic acid
`and sodium sulfite, respectively], when compared to normal nails
`(iontophoresis). Interestingly, the amount permeated using ionto-
`keratolysis followed by iontophoretic delivery of TH was ∼4-fold
`higher than the passive permeation in nail without the dorsal layer
`(abraded nail).
`
`3.2. Drug load in nail
`
`3.2.1. Passive delivery
`It is well-known that during the process of permeation across
`a tissue barrier, some amount of drug is retained in the barrier,
`and the amount of drug retained is likely to be influenced by the
`thickness of the barrier. Thus, it was hypothesized that the amount
`of drug loaded into the nail would be significant given the thick-
`ness of the nail. Further, it was hypothesized that the drug loaded
`in the nail would release over time such that the nail bed and sur-
`rounding tissues would be bathed in drug for an extended period of
`time, which would also prolong the therapeutic effect. Fig. 2 com-
`pares the drug load in the active diffusion area of the nail using
`both passive and iontophoretic delivery with nails subjected to
`various pretreatments. The drug loaded in the dorsal/ventral layer
`abraded, defatted and keratolysed nails was comparable to the nor-
`mal nail on a per-weight basis (mg/g). However, nails subjected
`to ionto-keratolysis (1 h) with salicylic acid or sodium sulfite fol-
`lowed by passive delivery (24 h) showed a significantly higher drug
`load (∼2.5-fold) (P = 0.0014) in comparison to the control (passive
`untreated nails).
`
`3.2.2. Iontophoretic delivery
`The utility of iontophoresis to enhance the drug load in nail by
`driving molecules into the deeper nail layers, as opposed to passive
`delivery, was reported in our earlier study (Nair et al., 2009; Nair
`et al., in press). A similar trend was observed in the current study
`as well. Iontophoretic drug delivery resulted in a large amount of
`drug being loaded into the nail plate (3–10-fold) when compared to
`passive delivery, irrespective of the nail treatments (Fig. 2). A signif-
`icant enhancement in drug load (∼3-fold) was found in the dorsal
`layer abraded nails (0.64± 0.18 ␮g/mg, P = 0.0045), when compared
`to normal nails loaded using iontophoresis (0.20± 0.08 ␮g/mg). In
`contrast, no significant difference in drug load was observed when
`the ventral nail layer was abraded (0.24± 0.03 ␮g/mg) or the nail
`which was subjected to defatting (0.25± 0.05 ␮g/mg) or keratolysis
`(0.22± 0.09 ␮g/mg) processes (Fig. 2). However, ionto-keratolysis
`for a short duration (1 h) with salicylic acid or sodium sulfite fol-
`lowed by anodal iontophoresis (24 h) of TH enhanced the drug load
`in the nail plate by ∼2-fold (P = 0.0254) and ∼3-fold (P = 0.0063)
`respectively, when compared to iontophoresis with the untreated
`
`Table 1
`Amount of terbinafine loaded in the peripheral nail areas during passive and ion-
`tophoresis processes with different nail treatments.
`
`Nail treatments
`
`Amount in peripheral area (␮g/mg)
`
`Normal nails
`Dorsal layer abraded
`Ventral layer abraded
`Defatted nails
`Keratolysed nail
`Ionto-keratolysis with salicylic acid
`Ionto-keratolysis with sodium sulfite
`
`Control
`0.012 ± 0.006
`0.009 ± 0.004
`0.011 ± 0.006
`0.013 ± 0.008
`0.014 ± 0.005
`0.017 ± 0.005
`0.021 ± 0.006
`
`Iontophoresis
`0.031 ± 0.018
`0.097 ± 0.032
`0.054 ± 0.019
`0.042 ± 0.025
`0.046 ± 0.009
`0.069 ± 0.011
`0.080 ± 0.037
`
`nail. The results obtained further validate the capability of catho-
`dal iontophoresis to drive the keratolytic agent into the nail. The
`enhancement in drug load could be due to an increase in nail ker-
`atin binding sites (due to greater surface area) and/or creation of
`pockets for drug retention by the keratolytic agents when they are
`driven into the nail matrix using iontophoresis.
`Generally the whole nail apparatus, including the nail that is not
`accessible to topical treatment due to the proximal and lateral nail
`folds which overlap the nail, will be affected in nail diseases. The
`drug loaded in the active diffusion area represents the area of nail
`that is in direct contact with the formulation. In vivo, this would
`be the exposed part of the nail available for application of the drug
`formulation. Therefore, for optimal efficacy the drug loaded in the
`active diffusion area must diffuse laterally into the area that is not
`exposed to the formulation. In the present experiments, the amount
`of drug loaded into the nail surrounding the active diffusion area
`was also assessed. The amount of drug reaching the peripheral nail
`area depends on the concentration gradient between the active dif-
`fusion area and peripheral nail area. In the case of iontophoresis, it
`also depends on the current flow into the peripheral tissue regions.
`Table 1 summarizes the amount of terbinafine loaded in the periph-
`eral region of the nail during various treatments. The peripheral
`drug load was higher with iontophoresis when compared to pas-
`sive delivery. The amount of drug present in the peripheral nail is
`still well above the MIC (∼0.001–0.01 ␮g/mL) (Darkes et al., 2003)
`of terbinafine for dermatophytes. The drug load in the peripheral
`area also followed a similar trend that was observed in the active
`diffusion area with the different treatments.
`
`4. Conclusion
`
`Permeation studies using dorsal/ventral layer abraded nails
`indicated that the dorsal (upper) layer is the primary barrier in
`topical nail delivery. Permeation of TH through and drug load in
`the nail subjected to various treatments showed similar trends in
`both passive and iontophoretic delivery, although the extent of drug
`delivery varied with treatment. Iontophoresis, being an active deliv-
`ery process, enhanced TH delivery in all the cases. Permeation was
`not altered in defatted nails, suggesting that the lipid component of
`the nail is not a significant barrier to terbinafine penetration. How-
`ever, enhanced drug delivery was observed when the keratolytic
`agents were driven deep into the nail matrix using iontophore-
`sis (ionto-keratolysis) or the dorsal layer was abraded. This study
`demonstrates that chemical alteration (ionto-keratolysis using sali-
`cylic acid or sodium sulfite) or physical alteration (abrasion of dorsal
`nail layer-debridement) could be effective pretreatment methods
`for enhancing topical trans-ungual delivery.
`
`References
`
`Alberti, I., Kalia, Y.N., Naik, A., Bonny, J., Guy, R.H., 2001. Effect of ethanol and isopropyl
`myristate on the availability of topical terbinafine in human stratum corneum,
`in vivo. Int. J. Pharm. 219, 11–19.
`Baran, R., Kaoukhov, A., 2005. Topical antifungal drugs for the treatment of ony-
`chomycosis: an overview of current strategies for monotherapy and combination
`therapy. J. Eur. Acad. Dermatol. Venerol. 19, 21–29.
`Baran, R., Hay, R.J., Garduno, J.I., 2008. Review of antifungal therapy and the severity
`index for assessing onychomycosis. Part I. J. Dermatol. Treat. 19, 72–81.
`Batheja, P., Thakur, R., Michniak, B., 2006. Transdermal iontophoresis. Expert. Opin.
`Drug Deliv. 3, 127–138.
`Darkes, M.J.M., Scott, L.J., Goa, K.L., 2003. Terbinafine: a review of its use in ony-
`chomycosis in adults. Am. J. Clin. Dermatol. 4, 39–65.
`De Berker, D.A.R., Andre, J., Baran, R., 2007. Nail biology and nail science. Int. J. Cosmet.
`Sci. 29, 241–275.
`Di Chiacchio, N., Kadunc, B.V