Hydrogel increases diclofenac skin permeation and absorption

Abstract Purpose Topical nonsteroidal anti‐inflammatory drug formulations are used commonly to treat musculoskeletal pain and inflammation. Drug properties and formulation composition are the primary determinants of the transdermal drug delivery rate. The ex vivo transdermal flux through human skin of three topical diclofenac formulations was compared. Methods The formulations tested were hydrogel 1% diclofenac sodium and two emulsion gels (1.16%/2.32% diclofenac diethylamine, equivalent to 1%/2% diclofenac sodium). Human abdominal skin obtained during unrelated surgical procedures was stored at −20 °C until use. Skin specimens were thawed, prepared and placed in Franz diffusion cells (stratum corneum facing donor cell). The test formulation (~200 mg) was applied to the donor cell skin surface, and the receptor compartment was periodically sampled over 48 hours. The drug concentration in the receptor medium was determined by a validated HPLC method. Raman spectral imaging was performed to visualize the location and distribution of diclofenac. Results After 5 hours, the cumulative amount of hydrogel diclofenac transiting the skin was about 10 times that of the emulsion gel 1.16% (P=0.0004) and about twice that of the emulsion gel 2.32% (P=0.022). Similar results were seen after 9 hours. Raman spectroscopy showed that the hydrogel formulation was a homogeneous mixture of its various components, including diclofenac. The emulsion gels were non‐homogeneous, with diclofenac in close proximity to the lipophilic (paraffin) phase. Conclusions The transdermal transit of diclofenac from the hydrogel demonstrated a faster onset and a greater absorption rate than either emulsion gel formulation, suggesting that the hydrogel formulation may have a faster onset of action in underlying tissues vs. the emulsion gel products.

of 27.8% in those countries, followed by ibuprofen (11.0%) and naproxen (9.4%) (McGettigan & Henry, 2013). Similarly, diclofenac is used widely in the United States, where more than 10.8 million prescriptions for diclofenac products were written in 2012 (Iroko Pharmaceuticals, 2014). Diclofenac is available in several types of dosing forms, including oral and injectable forms for systemic dosing and topical products for local treatment of underlying tissues (Altman, et at., 2015).
Topically applied NSAIDs can produce clinically effective drug concentrations at the site of action in the underlying tissue while reducing systemic exposure (Brunner et al., 2005;Miyatake, et al., 2009), which reduces the risk of systemic side effects and could improve patient compliance. The challenge in topical transdermal drug delivery is to create a formulation that allows the drug to permeate quickly and efficiently through the stratum corneum, the outermost layer of skin that provides the barrier function of the organ (Haftek, et at., 1998). The stratum corneum consists of 10-20 layers of cornified cells embedded in a hydrophobic lipid-protein matrix (Haftek, 2015). The primary pathway for absorption of topically applied drugs through the stratum corneum is thought to be this intercellular matrix (Vitorino, et al., 2015). Factors that influence the transdermal permeation of topically applied NSAID formulations include chemical structure and properties of the drug (e.g. molecular weight, hydrophilic vs. hydrophobic properties, free-acid or free-base vs. salt form of the molecule), and the composition of the formulation, particularly the inclusion of excipients that enhance dermal penetration (Brunner et al., 2011;Escribano, et al., 2003;Folzer, et al., 2014;Lane, 2013;Marwah, et al., 2016;Nivsarkar, et al., 2015;Vitorino et al., 2015).
The onset of a beneficial effect of an NSAID upon topical application is influenced by the time lag before the active drug is absorbed and the rate of flux of the drug through the skin. Here the results are presented of ex vivo testing of skin permeation using a wellestablished method (Raney, et al., 2015) to compare three formulations of diclofenac and demonstrate that the hydrogel-based formulation delivers diclofenac more quickly and at a higher rate than either of the emulsion gel formulations. In addition, Raman spectroscopy suggested differences in the location of diclofenac within the lipophilic and hydrophilic phases of each of the formulations that may contribute to the differences in transdermal drug transport.

All experimental methods were performed using Good Laboratory
Practice and all instruments used in the analyses described below conformed to this.  (Novartis, 2013a(Novartis, , 2013b. The composition of each formulation is presented in Table 1. The active ingredient in the hydrogel 1% formulation is diclofenac sodium (10 mg/g), whereas the emulsion gel formulations use diclofenac diethylamine 1.16% or 2.32% (equivalent to 10 or 20 mg/g diclofenac sodium, respectively).

| Diclofenac for topical administration formulations
The USP diclofenac sodium was purchased from LGC Ltd (Teddington, UK) and was used as the standard for all analytic testing.

| Permeation and penetration into and across human skin
This ex vivo absorption method was based on the Organisation for Economic Cooperation and Development (OECD) Test No. 428: Skin absorption: in vitro method (OECD, 2004). Human abdominal skin samples from three patients were obtained via surgical skin removal procedures that were unrelated to the present investigation. Each of the three patients consented to the scientific use of skin prior to surgery.
Skin was not used if a pathological finding was present or if there was skin damage, strongly marked scarring or pregnancy stretch marks.
The excised skin was cooled to 4°C, the subcutaneous fatty layer was separated from the skin, and the skin specimen was stored at −20°C until use. To prepare the skin sample for use, the specimen was thawed and cut into strips with a scalpel and dermatomized to a mean thickness of 500 ± 100 μm, leaving the stratum corneum intact. The suitability of skin from each donor was assessed by measuring the transdermal transport of caffeine to confirm that the skin samples remained relatively impermeable to this low permeability marker. This confirmation was performed using a Franz cell as described below. The caffeine solution (10 mg/ml) was added to the donor cell and 320 μl samples were withdrawn from the receiver compartment at 4,6,8,20,24,28,32,46 and 48 hours. The sampled volume was replaced with phosphate-buffered saline (PBS) after each sample was removed.

| Franz cell
Circular samples from each skin specimen were prepared with a hollow punch and placed between the donor and receiver sections of a  (Khalil, et al., 2000).
At the conclusion of the permeation study, each skin sample was removed from the Franz cell and the residual drug delivery formulation was removed by gently swabbing with cotton swabs. The stratum corneum was removed by tape stripping as described by Wagner, et al., (2002). Each tape strip was placed in a vial for extraction of diclofenac or caffeine. Following tape stripping the skin sample was frozen at −80°C in a cryomicrotome. Parallel 25 μm thick sections were cut and placed in a single vial for the extraction of diclofenac or caffeine.

| Analytical methods
The extraction medium for diclofenac or caffeine was a 1:1 (v/v) mixture of ethanol and water. The tape strips were extracted in a minimum of 2 ml of medium and skin slices were extracted in 2 ml

| Calculations
The apparent permeation coefficient was calculated using Equation (1): Where P app is the apparent permeation coefficient (cm•s −1 ), dQ/dt is the steady state transport rate obtained by linear regression of the amount of test substance transported vs. time (μg/s), m 0 is the initial mass of test substance in the donor compartment (μg), A is the area of exposed skin (cm 2 ) and V D is the donor volume (cm 3 ).

| Raman spectral imaging
Raman spectral imaging (Smith & Dent, 2005) was performed to assess the location and distribution of diclofenac in the three formulations; testing was performed by WITec (Wissenschaftliche Instrumente und Technologie GmbH, Ulm, Germany). Samples were prepared and placed between two coverslips and excited with a 532 nm diode laser.

| Raman spectroscopy
Colour-coded Raman spectroscopy images of hydrogel 1%, emulsion gel 1% and emulsion gel 2% are presented in Figure 3. Important differences were seen in the hydrogel and emulsion gel formulations.
The hydrogel 1% appeared as a homogeneous mixture of its various components, including diclofenac. In contrast, the Raman spectroscopy image of emulsion gel 1% revealed a non-homogeneous mixture in which diclofenac was located in close proximity to the lipophilic phase (paraffin). The emulsion gel 2% sample also appeared to show diclofenac in proximity to paraffin.

| DISCUSSION
The major finding of this ex vivo transdermal drug delivery study is that diclofenac from hydrogel 1% applied to human skin samples was more quickly transported across the skin than was diclofenac from emulsion gel 1% or 2%. This difference was clearly demonstrated by the mean P app of diclofenac delivered from hydrogel 1%, which was 2.8-fold greater and nearly 2.1-fold greater than from emulsion gel 1% and emulsion gel 2%, respectively. Similar differences were also seen in the total amounts absorbed (Table 2). Importantly, this separation in efficiency of transdermal absorption was most evident after 5 hours, when 10 times more diclofenac had been transported across the skin from hydrogel 1% than from emulsion gel 1% (P=0.0004). The 5 hour time frame also revealed a substantial lag in the release and transdermal transport of diclofenac from the emulsion gel formulations. The faster onset and increased transdermal absorption rate of hydrogel 1% compared with the emulsion gel formulations suggest that hydrogel 1% would have a faster onset of action in underlying tissues compared with the emulsion gel products. The drug delivery advantage of hydrogel 1% was also evident at 9 hours, which is important because most topical NSAIDs are indicated to be applied 2 to 4 times daily (or about every 4 to 8 hours) (Novartis 2013a(Novartis , 2013b. The difference in transdermal delivery of diclofenac from hydrogel 1% and the two emulsion gel formulations observed in the ex vivo experiments described here is also likely to be seen during in vivo topical administration. In one of the original descriptions of this ex vivo method, Franz used his diffusion cell to compare the ex vivo human skin transit with in vivo transdermal absorption of a series of 12 randomly chosen organic compounds (Franz, 1975). Analysis of the total amount absorbed in the two tests revealed a significant rank-order correlation (correlation coefficient = 0.734; p < 0.01). Further development of the method and harmonisation of its protocol has provided support for its use to establish bioequivalence between transdermal pharmaceutical formulations (Abd et al., 2016;Franz, et at., 2009).  In order for diclofenac to be delivered to subcutaneous target tissues from a topically applied formulation it must release from the formulation into and diffuse through the lipophilic stratum corneum, through the underlying less lipophilic viable epidermis and ultimately through the dermis (Vitorino et al., 2015). The rate-limiting step is the partitioning into the stratum corneum, which is in part influenced by the relative solubility of the drug in the formulation and in the stratum corneum (Vitorino et al., 2015). The overall rate of transdermal transport is a function of the diffusion coefficient of the drug in each compartment or layer as well as physical factors, including the viscosity and the length of the diffusion path (Vitorino et al., 2015). Importantly, the physical and chemical composition of the stratum corneum creates a barrier with anisotropic diffusion properties featuring lateral diffusion coefficients 40 to 300 times the transdermal diffusion coefficients (Nitsche, et al., 2019). In addition, diffusion through hair follicles may be an important transdermal pathway, especially for molecules with low lipid solubility (Barbero & Frasch, 2017).
In the present study, diclofenac more readily penetrated and transited the skin from the hydrogel formulation than from either of the emulsion formulations. One possible explanation for this difference was suggested by the Raman spectral analysis. In the hydrogel formulation diclofenac appears to exist in solution, whereas in the emulsion gel formulation diclofenac diethylamine resides as particles in the lipid phase of an oil-in-water emulsion within a polyacrylatebased gel matrix. Thus diclofenac in the hydrogel is immediately available for diffusion into the skin compared with the emulsion gel in which diclofenac must first release from the lipid phase to be available for penetration of the skin (Seth, 1992). This concept is supported by work reported by Seth (1992), who compared the in vivo transdermal absorption of diclofenac sodium from a solution gel (an early developmental formulation of hydrogel 1%) and an emulsion gel (identified by the cited reference as diclofenac diethylamine 1.16%, Voltaren® FIGURE 2 Analyses of cumulative transdermal transport of diclofenac from three formulations designed for topical administration. Each value represents the amount of diclofenac transported through the skin into the receptor medium. Statistically significant differences are shown by brackets (Mann-Whitney U test). *p < 0.05; ***p < 0.001 Values are presented as mean ± SD.
Emulgel) in healthy volunteers. Diclofenac released from the solution gel demonstrated faster absorption as assessed by t max , 2-fold higher peak plasma concentration and nearly a 2-fold higher area under the curve (AUC) corrected for dose than diclofenac delivered from the emulsion gel. In addition, the water content of the hydrogel may increase hydration of the stratum corneum, which has been reported to cause morphological changes consistent with increasing the permeability of this barrier (Haftek et al., 1998;Marwah et al., 2016).
Increased epidermal hydration is also reported to increase the permeability of the skin; for review see Marwah et al. (2016).
As illustrated in the present study, differences in the compositions of topical diclofenac formulations have been shown to be the primary determinants of differences in the rate of transdermal drug delivery (Brunner et al., 2011;Escribano et al., 2003;Folzer et al., 2014;Nivsarkar et al., 2015). shown by plasma AUC and by underlying muscle concentrations (Okuyama et al., 1999). Others have reported that formulating diclofenac as an acid rather than as a salt may enhance the transdermal absorption compared with diclofenac emulsion gel (Brunner et al., 2011). In addition, small differences in ex vivo transdermal permeation rates have been reported for different salts of diclofenac when tested in simple solutions (Minghetti, et al., 2007), but whether these differences occur in complex transdermal formulations has not been described.
Although oral NSAIDs remain a first-line therapy for pain and inflammation of musculoskeletal disorders, including osteoarthritis, tolerability issues and safety concerns may limit their use in some patients (Henry & McGettigan, 2003;McGettigan & Henry, 2006  Raman spectroscopy. Visible images (left) and colourcoded images (right) from Raman spectroscopy of the three diclofenac formulations. The Raman colour-coded image of hydrogel 1% shows a homogeneous mixture of all formulation components. The emulsion gel 1% diclofenac, which is coded as red, appears primarily in association with paraffin, which is coded as yellow. A similar distribution of diclofenac is seen in the emulsion gel 2% in which the formulation remains unoccluded and subject to evaporation and erosion by clothing.

| CONCLUSIONS
The results of this study indicate that hydrogel 1% was superior to emulsion gel 1% over 48 hours or emulsion gel 2%, particularly over the first 9 hours, in its ability to effectively deliver diclofenac into and through human skin ex vivo. The Raman spectral analysis suggests that differences in the partitioning of diclofenac within these formulations may in part explain the differences in the absorption rate.
Although the skin absorption model is widely accepted for comparison of transdermal drug permeation, in vivo testing would be necessary to demonstrate that the ex vivo results translate to clinical use.

DATA AVAILABILITY
Data are available upon request from the corresponding author.