BI-LAMINATED ORAL DISINTEGRATING FILM FOR DUAL DELIVERY OF PITAVASTATIN CALCIUM AND LORNOXICAM: FABRICATION, CHARACTERIZATION AND PHARMACOKINETIC STUDY

Objective: Pitavastatin calcium (PT) is an innovative drug of statins that enhances HDL-C and lowers LDL-C. However, myalgia has been reported in hyperlipidemic patients receiving statins. Therefore, co-administration of statins with NSAIDs such as Lornoxicam (LN) could be a solution to the former problem. Accordingly, this study aimed to formulate a bi-laminated oral disintegrating film (ODF) comprising PT in one layer and LN in the second one. Methods: For the formulation and optimization of PT-ODFs, a 3 1 .2 1 full factorial design was carried out, where the impact of polymer type and concentration on disintegration time (DT) and % PT released after 10 min (Q 10 ) was studied. PT-ODFs were prepared via the solvent casting method and then evaluated. One PT-ODF was chosen to represent the optimum formula according to the criteria of scoring the fastest DT and the highest Q 10 . The optimized PT-ODF was merged with the second film layer containing LN, forming a bi-laminated ODF named S1 that underwent an in vivo pharmacokinetic study compared to the commercially available tablets for PT (Lipidalon ® ) and LN (Lornoxicam ® ) using rats as an animal model. LC-MS/MS was used to analyze plasma drug concentrations. Results: All PT-ODFs showed acceptable outcomes. F1 scored the fastest DT (18.6±1.5 s) and the highest Q 10 (91.3±3.0 %). S1 successfully recorded a maximum plasma concentration (C max ) of 2.04 and 2.24 folds increase for PT and LN, correspondingly, compared to commercially available tablets. Conclusion: Merging PT and LN into bi-laminated ODF was promising for the fast delivery of both drugs with enhanced bioavailability.


INTRODUCTION
Statins are a class of drugs that act against low-density lipoprotein cholesterol (LDL-C), triglycerides and help to enhance high-density lipoprotein cholesterol (HDL-C) in the blood, such as pitavastatin calcium (PT), which is categorized as a novel drug of statins. However, discontinuation of statins is commonly detected among patients due to the incidence of numerous side effects, the most common of which is myalgia [1][2][3][4]. Therefore, co-administration of nonsteroidal anti-inflammatory drugs (NSAIDs) with statins can support the treatment of the previously mentioned side effects [5]. NSAIDs are commonly used in the management of muscle pain [6]. One of these NSAIDs is lornoxicam (LN), which is a potent painkiller and anti-inflammatory drug [7,8]. Among different administration routes, the oral route continues to be the most effective because it is easy to use, non-invasive, provides an accurate dose and has high patient acceptance. Nevertheless, the incidence of low bioavailability due to the first-pass metabolism is still problematic. So, a novel oral drug delivery system was introduced, which is the oral disintegrating film (ODF), where the oral transmucosal route has attracted special attention as it merged the benefits of the oral route and systemic drug delivery without the first-pass effect [9,10]. This administration route is appropriate for use, particularly in elderly patients, as they struggle to swallow conventional tablets [11]. The ODF is a thin film that adheres to the tongue of the patient and dissolves instantly via saliva, releasing the drug for oro-mucosal absorption. So, using ODF would provide the advantages of fast onset of action accurate dosing, and it didn't need water for administration, which in turn would improve patient compliance [10,[12][13][14].
The purpose of this study was to formulate, evaluate, and optimize different PT-ODFs. Formulation variables affecting PT-ODF properties were studied by applying a 3 1 .2 1 full factorial design via Design-Expert ® software version 12. The PT-ODF that was chosen to exemplify the optimum formula was merged with the second film layer containing LN, forming a bi-laminated ODF for controlling myalgia triggered by the overuse of statins. An in vivo study in rats was carried out to appraise the pharmacokinetic parameters and relative bioavailability of both drugs (PT and LN) in the bi-laminated ODF compared to the commercially available tablets.

Formulation of PT-ODFs
The PT-ODFs were prepared using different polymers (pullulan, PVA, or HPMC) via the solvent casting method, where pullulan and HPMC were accurately weighed and soaked in cold distilled water, while PVA was dissolved at 80 °Cin distilled water to form a polymeric solution. Then PT, glycerol, and orange flavor were dissolved in a mixture of ethanol and distilled water (1:1) and added to the previously prepared polymeric solution and stirred till a homogenous solution was formed. The solution was degassed in a sonicator to remove air bubbles, then 25 ml of the solution was transferred into a petri-dish with a diameter of 9.8 cm. To avoid sudden evaporation of the solvent and reduce blistering of the film surface, a funnel was inverted above the petri dish after pouring. The solvent was allowed to evaporate for 24 h for film formation. Then the film was removed and cut to the desired size (2x2 cm) using a sharp razor blade, then enfolded in aluminium foil and left in a desiccator till evaluation [12,15].

Visual examination
All prepared PT-ODFs were assessed visually for their color, transparency, smoothness, and softness [16].

Thickness
The thickness of the film was determined via a micrometer at five different locations (the center and four corners). The average and standard deviation (SD) were calculated [17][18][19].

Surface pH
The film was wetted with 0.5 ml of simulated saliva fluid (SSF, pH 6.8) and left for 30 s in a petri dish. The pH was noted via a pH meter. The readings were reported as mean±SD [17].

Drug content
The film was dissolved in SSF (pH 6.8) in a 100 ml volumetric flask. The absorbance of the solution was determined spectrophotometrically at λmax = 245 nm using a blank (SSF, pH 6.8) for the calculation of drug content [20].

Folding endurance
The value of folding endurance was assessed by folding the film at the same point till it broke or folding it up to 300 times [15,21,22].

In-vitro disintegration time
The test was done by placing the film in a petri dish containing 6 ml of SSF (pH 6.8). The time the film began to disintegrate was noted. The average and SD were calculated [17].

In vitro drug release study
The test was done via the USP dissolution system, Distek (Model 2500i Type II, TCS-0500 Scheduler, New Jersey, USA) at 37±0.5 °C and a paddle speed of 50 rpm. It was performed by adhering the film to a glass plate using an adhesive and immersing it in 100 ml of SSF (pH 6.8). At time intervals of 2, 4, 6, 8, 10, 20, and 30 min, aliquots of 5 ml were taken and exchanged with equal volumes of SSF (pH 6.8).
The absorbance was determined spectrophotometrically at λmax 245 nm and the % PT released was calculated [15].

Optimization of the prepared PT-ODFs
The results of in vitro DT and drug release were analyzed by Design Expert ® software version 12. Depending on the desirability constraints of factorial outcomes of dependent variables, DT and Q10. The PT-ODF that recorded the fastest DT and the highest Q10 with the highest desirability value was selected as the optimum formula.

Scanning electron microscopy (SEM) of the optimum PT-ODF
The topography and structure of the sectioned surface of the optimum PT-ODF were inspected via SEM. The film was transversely cut and the texture was investigated via SEM at a voltage of 20 kV. A thin part of the film was cut with a scalpel for the preparation of the cross-section sample [23,24].

Fourier transform infrared spectroscopy (FT-IR) study
Pure PT, the physical mixture of PT with excipients, pure LN, the physical mixture of PT and LN, and the physical mixture of the optimum PT-ODF with LN were all analyzed by FT-IR (Shimadzu ® , Japan) via the potassium bromide (KBr) pellet method. 3 mg of sample and 300 mg of KBr were grounded by means of a mortar and pestle. A small portion of the mixture was compressed at 10 kg/cm in a hydraulic press, resulting in a transparent pellet that was inserted into the sample holder and subjected to scanning in the range of 4000 to 500 cm -1 [5,24].

Second derivative spectrophotometric analysis for instantaneous determination of PT and LN in SSF (pH 6.8)
PT and LN mixtures in SSF (pH 6.8) were prepared. The concentration was detected by measuring the absorbance of each drug in different mixtures at PT and LN wavelengths. The recovery percent (R %) was calculated for each mixture, and the interference degree was identified [25].

Formulation of bi-laminated ODF containing PT and LN
The bi-laminated ODF comprises the optimum PT-ODF layer and the LN-ODF layer. Its composition is shown in table 2. Each layer was formulated separately via the solvent casting technique, as mentioned before. One of the two layers was sprayed with a few drops of the casting solvent, followed by pressing the second layer on the sprayed side, forming the bi-laminated ODF named S1, and then it was subjected to air drying for 24 h and checked for complete adhesion between both layers prior to use [21]. A preliminary trial was performed for the preparation of ODF containing both drugs (PT and LN), and it was found that combining two drugs into a single film affected the film elasticity as it showed a low folding endurance value. So, it was preferred to formulate a bi-laminated ODF containing PT in one layer and LN in the second one to ensure film elasticity. The bi-laminated ODF exhibited a folding endurance value of up to 300 folds, which indicated the film elasticity as the higher the folding endurance value, the more elastic the film [26,27].

Characterization of the bi-laminated ODF containing PT and LN
The characterization tests for the S1 ODF were carried out using the same methodologies as those used for PT-ODFs. The in vitro release of PT and LN from S1 ODF was compared to that of the commercially available tablets.

In vivo pharmacokinetic study of the bi-laminated ODF containing PT and LN
In this study, twelve male Wistar rats were used. They were haphazardly distributed into two groups, each with six animals weighing between 300-350 g. An ethics committee (PI 2651), Faculty of Pharmacy, Cairo University, Egypt, approved the study protocol. Using a parallel design, group I received S1 formula with a dose of 1 mg/kg [28,29] for PT and 1.6 mg/kg [30] for LN, and group II received the commercially available tablets with the same doses for PT (1 mg/kg) and LN (1.6 mg/kg) based on rat body weight. The commercially available tablets were grounded in distilled water [27] with 0.9 ml containing (0.3 mg) of PT and 1.2 ml containing (0.48 mg) of LN. The S1 ODF was cut into the desired size, containing the adjusted dose of PT and LN. Each group was accommodated in a well-ventilated cage with a controlled temperature environment, a 12 h light-dark cycle, and a regular laboratory diet with free water access [31]. Before the study, the rats fasted for 12 h. Their oral cavities were cleaned to eradicate any debris prior to placing the bilaminated ODF on the tongue by tweezers. To aid film disintegration, a slight amount of water (50-200 µl) was added [32]. The rats were fed 4 h after receiving the drug and had unlimited access to water throughout the study. Under light anesthesia by diethyl ether, blood samples (0.5 ml) were collected by a heparinized tube from the retro-orbital venous plexus [33] at time intervals of 0, 0.25, 0.5, 1, 2, 4, 8, 24, and 48 h after dosing. The blood samples were centrifuged instantly at 4000 rpm for 5 min (PLC-012, Gemmy Industrial Corp, Taiwan). Liquid Chromatography-Mass Spectrometry (LC-MS/MS) (Shimadzu ® , Japan) was used to analyze the plasma supernatant by a valid LC-MS/MS analysis method [24,34,35].

Sample preparation and LC-MS/MS assay of PT and LN
A100 µl of torsemide (internal standard) from a (200 ng/ml) stock solution was added to 0.5 ml rat plasma sample. 4 ml ethyl acetate was used to extract torsemide and plasma samples, which were vortexed for 2 min before being centrifuged at 5000 rpm for 5 min. The supernatant was filtered with a 0.22-µm membrane filter and it was evaporated till dryness via a vacuum concentrator. The dried residue was reconstituted in 0.5 ml mobile phase; then a 20 µl aliquot was loaded to LC-MS/MS system (Shimadzu ® , Japan) with a triple quadrupole detector API-3200, AB Sciex (Foster, CA, USA) to analyze the plasma concentrations. The isocratic flow rate was 1.0 ml/min. The mobile phase was made up of 80% acetonitrile and 20% 0.01% ammonium acetate. The C18 column (4.6×50 mm, 5 µm diameter; Agilent Zorbax, CA, USA) was used to separate the samples. Analyst software version 1.6 was used to process the analytical data [34].

Pharmacokinetic parameters analysis
The pharmacokinetic parameters of PT and LN in the S1 ODF and the commercially available tablets were analyzed using a noncompartmental model via Kinetica ® software version 5 [24]. The maximum values of PT and LN plasma concentration (Cmax), the time to reach Cmax (Tmax), half-life (T1/2), mean residence time (MRT), area under the curve (AUC0-48), and (AUC0-∞) were calculated. Also, the percent relative bioavailability (Frel%) of the S1 formula compared to the commercially available tablets was calculated:

Statistical analysis
The statistical software of the statistical package for social sciences (SPSS ® ) version 26 was used to analyze the outcomes of the characterization tests of PT-ODFs. To determine the significance among the prepared PT-ODFs at (*p<0.05) in terms of in vitro DT and drug release, a one-way analysis of variance (ANOVA) was done, followed by post hoc multiple comparisons using the least square difference (LSD). A parametric one-way ANOVA test was applied to reveal the significance or non-significance of the pharmacokinetic results of Cmax, AUC (0-48), AUC (0-∞) between the S1 ODF and the commercially available tablets. A non-parametric Kruskal-Wallis test was done to compare the Tmax data obtained from the S1 ODF and the commercially available tablets via SPSS ® version 26. The results are significant when (*p<0.05).

Characterization of PT-ODFs
The results of all prepared PT-ODFs are presented in table 3. The films were transparent, colorless, thin, and smooth. The thickness values of the PT-ODFs ranged from 0.13±0.05 mm for F1 to 0.26±0.05 mm for F6. All films showed folding endurance values of up to 300 folds, which indicates good film flexibility [15,21]. All films had a surface pH close to the salivary pH (6.8), ranging from 6.89±0.09 for F3 to 7.02±0.03 for F1. The assay of drug content of all PT-ODFs was within the compendia specifications as it lies in the range between 94.8±1.8 % for F4 and 100.2±0.9 % for F1, and this indicates the uniform distribution of the drug in the film [15].  fig. 1. The statistical analysis revealed that the polymer type (X1) exhibited a significant impact on DT (*p<0.05). Pullulan-based ODFs recorded the shortest DT [36]. This might be due to the fact that pullulan is a water-soluble polymer that erodes rapidly, forming pores filled with the solvent diffusing into the film, causing it to disintegrate faster [37,38]. Conversely, HPMC-based ODFs recorded the longest DT. This could be attributed to the fact that HPMC is a viscous film-forming polymer that promotes the formation of a thick gel fluid when the film is in contact with the medium, leading to disintegration hindrance [39]. Concerning polymer concentration (X2), it significantly influenced the DT (*p<0.05). ODFs containing 50 mg recorded a shorter DT than those containing 80 mg. This could be due to the fact that, as the polymer concentration increased, the film thickness also increased. As a result, the time taken for the disintegration of the film increased, and this was in harmony with Raza et al.; Esim et al. [40,41].  fig. 1. The statistical analysis showed that the polymer type (X1) significantly influenced Q10 (*p<0.05). Pullulan-based ODFs showed the highest Q10. As mentioned earlier, this might be credited to the fact that pullulan is a water-soluble polymer that increases wettability and water permeation into the film matrix, promoting the creation of pores and channels on the surface of the film, resulting in increased drug release [38,42]. While HPMC-based ODFs recorded the lowest Q10, this could be due to the viscous nature of HPMC, which promotes the creation of a thick gel fluid upon contact with the medium, as previously mentioned, resulting in a delay of drug release from the film [43]. Regarding polymer concentration (X2), it significantly influenced Q10 (*p<0.05). ODFs formulated using 50 mg showed a higher Q10 than those prepared using 80 mg. This might be due to that as the polymer concentration decreased, the % drug released increased, as the release of the drug would be less hindered by the polymeric matrix, and this was in accordance with Alsofany et al. [44].

Optimization of the prepared PT-ODFs
The Design Expert ® software version 12 was an effective way to choose the optimum formula. The software selected the F1 formula, which had the fastest DT (18.6±1.5 s) and the highest Q10 (91.3±3.0 %) with overall desirability of 0.985 as the optimum formula.

SEM of the optimum PT-ODF
The SEM of the cross-sectional view of the optimum PT-ODF (F1) is displayed in fig. 3. The micrograph showed that the film had large, deep, and diffusible pores. These pores formed channels through which the water penetrated, facilitating film disintegration and drug diffusion. This was confirmed by the previous outcomes of in vitro DT and drug release [45]. Additional peaks were noticed at 828.40 cm -1 (-CH) and 779.73 cm -1 (C-Cl) [5,34]. The main peaks of PT and LN in their physical mixture and in the S1 formula physical mixture persisted in their positions, representing the lack of incompatibility.
Second derivative spectrophotometric analysis for instantaneous determination of PT and LN in SSF (pH 6.8) At λmax 376 nm of LN, PT had no reading; it gave zero absorption. As a result, there was no interference for measuring LN in the presence of PT. Conversely, there was a reading for LN at λmax 245 of PT, indicating that there was overlapping between PT and LN as shown in fig. 5. To measure PT in the presence of LN, second derivative method was done. The absorbance and second derivative spectra were recorded for both drugs in SSF pH 6.8: A suitable wavelength was observed via the zero-crossing method. The spectra showed that the zero-crossing of LN was at 246.2 nm as shown in fig. 6. The R % was (95.1±0.3 % to 101.5±0.5 %) for PT and (96.7±0.5 % to 102.6±0.2 %) for LN. The interference degree was determined from the R% of the mixtures. There was no interference observed [5].  Table 3 displays the results of the characterization tests done on the S1 ODF, where it exhibited a folding endurance value of up to 300 folds which indicates the flexibility of the bi-laminated ODF. The drug content assay was within the compendia specifications. Fig. 7 shows the in vitro release profile of PT and LN from S1 and the commercially available tablets, where Q10 was 90.21±2.3 % and 83.06±2.4 % for PT and LN, correspondingly, in S1 ODF, and this was higher than the commercially available tablets that recorded Q10 of 54.58±1.9 % for PT and 47.72±2.17 % for LN.  Table 4 and fig. 8 show that the mean value of Cmax of the S1 ODF and Cmax of the commercially available tablets were significantly different (*p<0.05), with 2.04 and 2.24 folds increase regarding PT and LN, correspondingly, compared to the commercially available tablets. The statistical analysis revealed that the mean value of Tmax of S1 was significantly shorter (*p<0.05) than the commercially available tablets. The mean value of AUC (0-∞) of S1 was 2.3 times for PT and 2.14 times for LN, which was significantly (*p<0.05) greater than that of commercially available tablets. Additionally, the Frel% of S1 was 230.04 % for PT and 214.02 % for LN with respect to the commercially available tablets. The results of Tmax, AUC (0-∞), and Frel% showed that S1 ODF attained good sound results compared to the commercially available tablets. The enhanced bioavailability and the faster absorption rate of the two drugs, as marked by higher Cmax and faster Tmax compared to the commercially available tablets, might be due to that the drug was partially absorbed via the oral mucosa [12]. Additionally, the fast absorption of both drugs from S1 correlates well with the in vitro drug release results, where PT and LN showed rapid release from S1 ODF.   Data are presented as mean value±SD, n=6. * S1 is significantly different compared to the commercially available tablets at p<0.05. Cmax, maximum plasma concentration; Tmax, time to reach maximum plasma concentration; MRT, mean residence time; T1/2, half-life; AUC, area under the curve; Frel%: percent relative bioavailability; PT, pitavastatin calcium; LN, lornoxicam; ODF, oral disintegrating film.

CONCLUSION
A 3 1 .2 1 full factorial design was applied to choose the optimum PT-ODF. F1 prepared using pullulan at a concentration of 50 mg showed good sound results in terms of DT and Q10. Merging PT and LN was successfully done via the formulation of the bi-laminated ODF, comprising both drugs to support the treatment of myalgia caused by statin overuse. The in vivo pharmacokinetic study revealed that the bi-laminated ODF (S1) was found to be optimal for delivering the two drugs with enhanced bioavailability and a faster absorption rate with respect to the commercially available tablets. Nevertheless, further clinical pharmacokinetic and pharmacodynamic studies are needed to confirm the attained results and to verify that the coadministration of PT and LN can treat myalgia induced by statin overuse.

STATEMENT OF ANIMAL RIGHTS
The Institutional Animal Ethics Committee (PI 2651), Faculty of Pharmacy, Cairo University, Egypt, approved the study protocol.

AUTHORS CONTRIBUTIONS
All the authors have contributed equally.