1. Introduction
In recent years, conventional drug dosage forms for topical application (ointments, creams, etc.) have offered limited advancements with regard to (trans)dermal delivery. On the other hand, adherence to such therapy protocols is particularly low and often requires frequent and messy drug application [
1]. High-application frequency due to the low substantivity of semi-solid formulations, wherein up to 90% of the active substance is removed from the skin by clothing or by contact with other surfaces, negatively affects patient adherence and treatment efficacy [
2]. Therefore, film-forming systems (FFSs) have emerged as an esthetically acceptable option for targeted, less frequent and controlled drug delivery.
These alternative dosage forms defined as non-solid, i.e., intermediate between transdermal patches and semisolid dosage forms, produce a film after application on the skin. By maintaining a close contact with the skin for prolonged periods of time, FFSs enable less frequent dosing and improve patient compliance [
3].
Upon application, rapid evaporation of solvent(s) occurs, leaving a discrete in situ formed film on the skin surface, which is often imperceptible to the patient. Taking into consideration several aspects of prospective topical drug delivery systems (manufacturing costs, product quality, patient acceptability, etc.), FFS based on rationally selected film-forming polymers and conventional volatile solvents are expected to gradually increase their presence on the market [
4,
5].
Currently, several patented film-forming technologies based on polymers for dermal/transdermal drug delivery, such as Liqui-Patch
® (Epinamics GmbH, Berlin, Germany), Medspray
® (MedPharm Ltd., Guildford, UK) and Durapeel
® (Crescita Therapeutics Inc., Mississauga, ON, Canada) are available on the market [
6,
7,
8]. Evamist
® MTDS was the first metered-dose transdermal spray developed by Acrux Inc. (Melbourne, Australia), further commercialized in 2016 under the Lenzetto
® brand, for transdermal delivery of estradiol in the treatment of hot flushes commonly associated with menopause [
9]. Another product developed by Acrux, Axiron
® topical solution (Eli Lilly and Co., Indianapolis, USA), was the first sprayable film-forming solution approved by the FDA in 2010 for transdermal delivery of testosterone, which is used for hormonal replacement therapy in male hypogonadism (congenital or acquired) [
10]. Another film-forming cutaneous solution present on the market is Lamisil Once
® 1% (GlaxoSmithKline Consumer Healthcare, Brentford, UK), which contains the antimycotic drug terbinafine hydrochloride for the treatment of athlete’s foot (
tinea pedis) with a single application [
11].
However, a more systematic approach to FFS assessment is essential to consistently deliver the intended performance of the product, ultimately leading to more marketed products.
The dynamic nature of FFS implies a need to define a tailored quality target product profile (QTPP), which can only be achieved by smart identification of critical quality attributes (CQAs) so that those product characteristics that have an impact on product quality can be further studied and controlled [
12,
13]. Apart from standard aspects relating to quality, safety and efficacy of preparations for cutaneous application, the so-called metamorphosis/transformation of FFS deserves careful assessment, considering its prospective influence on a number of properties (e.g., precipitation of the incorporated active pharmaceutical ingredient). Although the importance of testing metamorphosis is acknowledged by the latest EMA draft guideline [
14], no single method appears to sufficiently characterize the entire process. Hence, this study encompassed a number of characterization methods, which were performed at four levels whenever applicable: with drug-loaded and unloaded FFS, in liquid form and upon film formation.
After assessing fundamental physicochemical, mechanical and tribological properties further reinforced with biocompatible and biosubstantive features of a series of FFSs [
15], the refined number of samples was selected to enter the next stage of the study, aiming to assess precise film morphology and potential molecular interactions within such dynamic system. The dipropionate ester of betamethasone, herein chosen as the model active substance, is a potent corticosteroid frequently used in treatments of different types of dermatitis and chronic plaque psoriasis localized on elbows, knees, scalp and feet [
16,
17,
18]. Betamethasone efficacy in the abovementioned indications might be improved with modern formulation strategies, such as targeted application of the polymeric FFS with enhanced substantivity at the target site, e.g., psoriasis plaques or the flexor surfaces of the forearms and legs. Furthermore, possibly reducing the application frequency of potent local corticosteroids and, subsequently, their undesirable effects, such systems could have a remarkable effect on patient compliance. Therefore, with this work, we aimed to contribute to the biophysical elucidation of the mechanism behind the FFS-to-film transition, possibly leading to enhanced drug dermal delivery of a lipophilic model compound—betamethasone dipropionate (BDP). After applying a set of physicomechanical methods (revealing the spreadability and drying time of the FFS, as well as the thickness of the generated films, their flexibility and basic sensory properties), thermal (differential scanning calorimetry, DSC) and spectroscopic (Fourier Transform Infrared Spectroscopy, FT-IR) methods were used as tools for the analysis of profound interactions that might occur upon FFS drying. Atomic force microscopy (AFM) was important for in depth film topography characterization and detection of potential drug crystallization during the phase transitions, whereas FFS viscosity and pH were monitored as a measure of preliminary stability during 6 months of storage. Because the drying process may cause phase transitions that affect not only physicochemical properties but also dermal availability [
5] in the final stage, an in vitro permeation study was performed to compare the selected FFS and a commercial product comprising the same concentration of the lipophilic model drug, BDP.
3. Materials and Methods
3.1. Materials
Betamethasone dipropionate (BDP), the model active pharmaceutical ingredient, was kindly donated by Galenika (Belgrade, Serbia). The drug reference product used for in vitro permeation study was Beloderm® ointment, purchased from Belupo (Zagreb, Croatia). Hydrophilic polymer Klucel® GF (hydroxypropyl cellulose) was supplied by Caesar & Loretz GmbH (Hilden, Germany). Hydrophobic polymethacrylate copolymers Eudragit® NE 30D (poly(ethyl acrylate-co-methyl methacrylate) 2:1 (polyacrylate dispersion 30% Ph. Eur.)) and Eudragit® RS PO (poly(ethyl acrylate-co-methyl methacrylate-co-trimethylammonioethyl methacrylate chloride), i.e., copolymer type B, Ph. Eur.) were obtained from Evonik Rohm GmbH (Darmstadt, Germany). Polysorbate 80 (Sigma Aldrich, Schnelldorf, Germany), glycerol, propylene glycol and solvents used (all Carl Roth GmbH, Karlsruhe, Germany) were of pharmaceutical grade. For HPLC-MS/MS analysis, acetonitrile LC-MS grade (J.T. Baker, Phillipsburg, NJ, USA), formic acid (Acros Organics, Geel, Belgium) and deionized water (Gen Pure Ultrapure, Thermo Fisher Scientific GmbH, Dreieich, Hessen, Germany) were used.
3.2. Preparation Method of the Polymeric FFSs with a Lipophilic Model Drug
The samples were prepared according to the compositions specified in
Table 1. The content of BDP was set to suit its therapeutic concentration of 0.064%, corresponding to 0.05% betamethasone. In a closed vessel, BDP (previously measured on an ABJ 120–4 M analytical scale; Kern & Sohn GmbH, Balingen, Germany) was dissolved in a small amount of the selected solvent (ethanol 96% (
v/
v) or isopropyl alcohol), wherein the addition of plasticizer (propylene glycol) depended on the chosen type of the polymer (Eudragit
® RS, HPC or their combination) and, optionally, the addition of the surfactant/potential penetration enhancer (polysorbate 80), while stirring continuously until the drug was completely dissolved. The chosen polymer or a combination of polymers was gradually dissolved/dispersed in the residual amount of the selected solvent(s) and a smaller amount of water while stirring continuously with a magnetic stirrer (RCT basic, IKA, Staufen, Germany). After 2.5–3 h of mixing, the pre-dissolved drug was added to the polymer dispersion. The mixing was continued on the magnetic stirrer up to 24 h, until complete homogenization. All formulations were stored in tightly closed dark glass vials.
3.3. Characterization of Physicochemical, Mechanical and Sensory Properties
3.3.1. Assessment of Drying Time, Thickness, Spreadability, Flexibility and Stickiness of the Casted Films
In order to assess the full impact of BDP incorporation into FFS, the samples were subjected to a series of specifically optimized characterization methods in order to screen for the following properties of the generated films: drying time, thickness, spreadability, flexibility and basic sensory properties. This allowed for comparison of the obtained numerical and descriptive values with the results attributed to placebo samples published in our previous study [
15]. All the parameters were assessed in triplicate in order to express the results in the form of a mean value ± standard deviation (SD) whenever applicable.
Drying time of the films casted by 500 μL of each sample was measured with a stopwatch, both at room (23 ± 2 °C) and skin temperature (32 ± 0.1 °C; Orbital Shaker Incubator ES 20, Biosan, Riga, Latvia) by touching the film surface. The obtained mean drying times were expressed in min ± SD and graded as low (≤5 min), medium (5–7 min) or high (>7 min), according to [
31].
Film thickness was assessed with a digital micrometer compliant with DIN 863, supporting a range of thickness measurements of 0–25 mm/0.001 mm (Kern, Balingen, Germany), by applying 10 μL of the samples to a microscope glass plate. After complete drying, the thickness of the film was measured and corrected to the thickness of the clean substrate, resulting in mm ± SD.
Film spreadability was evaluated via film surface measurement by applying 10 μL of each sample perpendicularly to the surface of a microscope glass plate. After complete film drying, the diffusion area was determined using a millimeter graph paper and expressed in mm2 ± SD.
Flexibility (key attribute of a film’s mechanical resistance) was evaluated with the so-called folding technique. The folding endurance value is defined as the number of times a film can be folded at the same place without breaking. Hence, the lower the folding endurance value, the more brittle the film is, allowing the test to assume a film’s integrity [
26,
32]. A volume of 500 μL of each sample was uniformly distributed over a length of 4 cm onto a 27 × 4.4 cm rubber substrate. After the film dried, a rubber band was repeatedly rolled and unrolled over the whole length of the film. After each rolling cycle, the film was inspected for changes using a magnifying glass. The resulting number of folds (i.e., the folding endurance value) was expressed as the number of rolling times up to the first perceived, visible change of the film on the rubber substrate ±SD.
The visual appearance of the formed films was descriptively appraised in terms of color, transparency, structure, homogeneity, gloss and stickiness. Film stickiness was additionally evaluated by applying a piece of cotton bud to the dry film [
31]. The cotton bud was weighed on an analytical balance (ABJ 120–4 M, Kern & Sohn GmbH, Balingen, Germany) both before and after application, which allowed the stickiness to be rated as ‘low’ (no observable cotton fibers on the film surface), ‘medium’ (a thin layer of fiber filaments is visible, with 0.01 g as the cutoff value of the retained cotton) or ‘high’ (0.02 g as the cutoff value of the retained cotton).
3.3.2. pH Values of FFS
Measurement of FFS pH values was performed by direct immersion of the previously calibrated pH checker® HI98103 probe (Hanna Instruments, Woonsocket, RI, USA). Testing was carried out initially (24 h after sample preparation), as well as after 3 and 6 months of storage at room temperature, as a manner of stability screening. All the measurements were performed in triplicate, resulting in mean values of three individual measurements (±SD).
3.3.3. Viscosity Measurements
Rheological profiles of drug-loaded FFSs and their respective placebo samples were assessed with an Anton Paar MCR 302 rheometer with RheoCompass software (Anton Paar GmbH, Graz, Austria). A measuring tool for low-viscosity and viscoelastic liquids was used (Concentric cylinder CC27), and the measurements were performed at 20 ± 0.1 °C. Average viscosity at different shear rates was considered important for both sample characterization and stability evaluation. Due to the volatility of FFSs, as a manner of stability control, viscosity was assessed initially, as well as after 3 and 6 months of storage at room temperature.
3.4. Morphology and Methamorphosis Characterization
3.4.1. Differential Scanning Calorimetry (DSC)
After assessing thermal changes of BD and relevant excipients used for sample preparation (namely, film-forming polymers, polysorbate 80 and propylene glycol), DSC was performed on the samples. Considering the volatile nature of the investigated FFSs, measurements commenced after complete drying of the generated films (the samples were casted into aluminum pans and left overnight to allow for complete solvent evaporation). Depending on the nature of the formulation, 2 to 10 mg of the dried samples were heated from 25 to 220 °C, with a heating rate of 10 °C/min, using DSC 1 (Mettler-Toledo GmbH, Greifensee, Switzerland). The nitrogen flow was set to 50 mL/min.
3.4.2. Fourier Transform Infrared Spectroscopy (FT-IR) Analysis
The attenuated total reflectance FT-IR spectra were recorded using a Nicolet iS10 FT-IR spectrometer (Thermo Fisher Scientific, Horsham, UK) in the wavelength range between 4000 and 600 cm
−1 with a resolution of 4 cm
−1. After obtaining the spectra of all relevant substances (
Table 1), the samples were analyzed both in liquid and dried state in order to assess possible drug–matrix interactions, as well as transformation phenomena of the FFS. Peak matching was performed to detect any possible interactions between the components.
3.4.3. Atomic Force Microscopy
In order to investigate the 3D structure and the homogeneity of the polymeric film, AFM analysis of the samples was performed by applying an NTEGRA Prima atomic force microscope (NT-MDT, Moscow, Russia). Intermittent-contact AFM mode was applied using NT-MDT NSGO1 silicon, N-type, antimony doped cantilevers with Au reflective coating. The nominal force constant of the cantilevers was 5.1 N/m, whereas the cantilever driving frequency was around 150 kHz. AFM images were created and analyzed with Image Analysis 2.2.0 (NT-MDT) and Gwyddion 2.60 (Free and Open Source software, Czech Metrology Institute, Jihlava, Czechia) software. Prior to actual measurement, the samples were cast on mica plates and left overnight at 30 °C.
3.5. In Vitro Skin Permeation Study
In vitro experiments were conducted under infinite dosing conditions using modified Franz diffusion cells (Gauer Glas, D-Püttlingen, Germany) with an effective diffusion area of 2.01 cm
2 and a receptor volume of 12 mL, with a porcine ear epidermis (heat-separated) as a membrane. The epidermal membrane was carefully prepared according to the previously described protocol [
33]. In brief, fresh porcine ears obtained from a local abattoir immediately after slaughter were washed under cold running water, and skin of the external side was carefully excised using a scalpel. The isolated skin was then cleaned off with isotonic saline and cotton swabs, blotted dry with a soft tissue, wrapped in aluminum foil and stored at −20 °C (within one month). On the day of the experiment, after thawing at room temperature, the skin was punched to the discs with a diameter 25 mm. Subsequently, the skin pieces were immersed in water at 60 °C for 90 s, and the SC-viable epidermis layer was separated from the underlying dermis using forceps. The isolated epidermal sheets were transferred to Petri dishes filled with phosphate-buffered saline (PBS; pH = 7.4) until use (within 1 h).
The heat-separated epidermal membranes were mounted between donor and receptor chambers filled with degassed, pre-heated (32 °C) receptor medium (a mixture of PBS and ethanol (96%) at a ratio 50:50 v/v). Afterwards, receptor parts of the cells were immersed in a water bath at 32 ± 1 °C in order to equilibrate the membranes under constant magnet stirring at 500 rpm. After 30 min, 300 μL of each of investigated FFS formulation (S1A, S2A, S3A) was uniformly applied to the membrane surface. Commercial product betamethasone ointment, 0.5 mg/g (Beloderm® ointment, Belupo, Zagreb, Croatia) was used as a drug reference sample (1 g of each FFS/1 g of Beloderm® ointment contains 0.64 mg BDP). Drug permeation through the heat-separated porcine ear epidermis was observed over the 26 h period to achieve a steady state, as well as to avoid inevitable degradation of the skin barrier. The temperature (32 ± 1 °C) and stirring speed (500 rpm) were maintained constant. At seven previously defined time points (2 h, 4 h, 6 h, 20 h, 22 h, 24 h and 26 h), 600 μL aliquots were withdrawn and immediately replaced with an equal volume of fresh, pre-heated receptor medium in order to maintain sink conditions. BDP concentration in the sampled aliquots was then assayed by the HPLC–MS/MS method. In order to fully characterize the permeation process, the cumulative amount of BDP permeation per unit area (ng/cm2) was plotted against time (t), and then the permeation rate (steady state flux) was determined from the slope of the linear portion of the plot for each investigated formulation. Permeation coefficients (ng/cm2 h) were calculated by dividing permeation rates (ng/cm2 h) by the initial concentration of BDP in the vehicle (ng/mg).
3.6. HPLC-MS/MS
BDP in aliquots obtained during in vitro permeation study was determined by the HPLC-MS/MS method. A liquid chromatographic system, Accela 1000 (Thermo Fisher Scientific, San Jose, CA, USA), consisting of an auto sampler and quaternary pump, was used. Chromatographic separation was achieved using a ZORBAX Eclipse Plus C8 column (150 mm × 4.6 mm, 5 µm; Agilent Technologies, Santa Clara, CA, USA) at 40 °C. A mixture of acetonitrile and 0.1% formic acid (80:20, v/v) was used for isocratic elution at a flow rate of 500 µL/min. The total analysis time was 6.5 min. Mass analyses were conducted on a TSQ Quantum Access MAX triple-quadrupole spectrometer equipped with a heated electrospray ionization (HESI) source, with high-purity nitrogen as nebulizing gas. HESI source parameters were optimized by syringe infusion (20 µL/min) of BDP standard solution. HESI source and mass spectrometry parameters were as follows: spray voltage, 5000 V; vaporizer temperature, 350 °C; sheath gas pressure, 30 units; ion sweep gas, 0 units; auxiliary gas, 10 units; ion transfer capillary temperature, 250 °C; capillary offset, 35 units; tube lens offset, 136 units; skimmer offset, 36 units; scan width (m/z), 0.02; scan time, 200 ms. Mass spectrometry was used to detect specific ions for analyte identification. Detection of the ions was performed in the selected reaction monitoring (SRM) positive scan mode using a transition of m/z 527.3 → 453.2 (collision energy was 20 V). Xcalibur software (Thermo Fisher Scientific, San Jose, CA, USA) was used for data acquisition and processing.
3.7. Statistical Analysis
Where applicable, data are presented as mean ± SD. Statistical analysis was performed using Student’s t-test or one-way analysis of variance (ANOVA), followed by a Tukey post hoc test or by a nonparametric Kruskal–Wallis test and by a Mann–Whitney U test for pairwise comparisons, depending on the nature of the data. An assessment of the normality of data was carried out using the Shapiro–Wilk test. Statistical analyses were performed using the PASW Statistics software package, version 18.0 (SPSS Inc., Chicago, IL, USA). The level of significance was set to p < 0.05.
4. Conclusions
FFSs are dynamic dosage forms with challenging characterization. Considering the possible therapeutic advantages these systems may convey to a modern patient, FFS development and evaluation should be rationally defined. The present study revealed the utility of several methods able to refine the number of needed tests within the final QTPP. In this regard, apart from the viscosity and pH value, physicomechanical properties (namely, film drying time, surface, folding endurance and thickness) are identified as important CQAs of FFSs because they directly affect ease of administration, skin retention, product performance and patient acceptability. Coupling AFM, DSC and FT-IR proved to provide a deeper insight into the mechanisms of metamorphosis/transformation of FFSs (e.g., absence of drug precipitation upon film generation while providing drug reservoirs), enabling a reduction in the number of in vitro biopharmaceutical and in vivo skin performance tests.
Thermal and spectroscopic characterization implied that BDP remains stable throughout the liquid-to-film transition in the investigated systems, comprising hydrophobic and/or hydrophilic polymers. At the 2D level, AFM depicts film homogeneity, whereas a 3D view exposes in-depth film morphology in the context of, among other factors, the presence and positioning of drug reservoirs. DSC contributes to confirmation of drug solubility (i.e., absence of drug precipitation) in the generated polymeric film. Therefore, by coupling these methods, tailored FFSs can be efficiently characterized in an early stage of development. In our study, various characterization stages (e.g., FFS viscosity, film thickness, ultimately confirmed in our permeation study) implied that the sample with a combination of film-forming polymers (Eudragit® RS PO and HPC) showed potential for sustained drug delivery. On the other hand, the Eudragit® NE 30D-based FFS increased the rate and extent of BDP permeation. The observed enhancement could be connected with the presence of valleys (inclusions) discovered by AFM. Naturally, only clinical studies would reveal more details about the treatment efficacy of such systems targeting psoriatic lesions; however, establishing a drug reservoir in the residual polymeric film could enable both sustained and enhanced skin delivery.