- Email: [email protected]
Orodispersible films: Product transfer from lab-scale to continuous manufacturing
International Journal of Pharmaceutics 535 (2018) 285–292
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Research Paper
Orodispersible films: Product transfer from lab-scale to continuous manufacturing Yasmin Thabet, Joerg Breitkreutz
T
⁎
Heinrich Heine University Duesseldorf, Universitaetsstrasse 1, 40225, Duesseldorf, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Orodispersible films Oral thin strips Film preparations Continuous manufacturing Hydrochlorothiazide Scaling-up
Orodispersible films have been described as new beneficial dosage forms for special patient populations. Due to various production settings, different requirements on film formulations are required for non- continuous and continuous manufacturing. In this study, a continuous coating machine was qualified in regards of the process conditions for film compositions and their effects on the formed films. To investigate differences between both manufacturing processes, various film formulations of hydrochlorothiazide and hydroxypropylcellulose (HPC) or hydroxypropylmethycellulose (HPMC) as film formers were produced and the resulting films were characterized. The qualification of the continuously operating coating machine reveals no uniform heat distribution during drying. Coating solutions for continuous manufacturing should provide at least a dynamic viscosity of 1 Pa*s (wet film thickness of 500 μm, velocity of 15.9 cm/min). HPC films contain higher residuals of ethanol or acetone in bench-scale than in continuous production mode. Continuous production lead to lower drug content of the films. All continuously produced films disintegrate within less than 30 s. There are observed significant effects of the production process on the film characteristics. When transferring film manufacturing from lab-scale to continuous mode, film compositions, processing conditions and suitable characterization methods have to be carefully selected and adopted.
1. Introduction Orodispersible films (ODFs) are defined as “single- or multilayer sheets of suitable materials, to be placed in the mouth where they disperse rapidly” (Ph.Eur., 2014). This dosage form provides the advantage of fully flexible dosing for several patient populations by cutting the film into the desired size. Furthermore, the film does not have to be swallowed like a conventional tablet, but only placed onto the tongue where it disperses rapidly into smaller pieces, which could then easily be swallowed. Therefore, paediatrics or geriatrics are potential target populations, which often suffer from swallowing difficulties (Hoffmann et al., 2011; Preis, 2015). There are several production techniques for ODFs described in literature. Beside the infrequently used hot melt extrusion process (Jani and Patel, 2015), the most common production method is via solvent casting method (Buanz et al., 2015; Cilurzo et al., 2008). Here, the active pharmaceutical ingredient (API) is dissolved or dispersed in a polymer solution. Additional ingredients like plasticizers, taste masking agents or fillers may be necessary (Dixit and Puthli, 2009). Afterwards, the solution is cast with a defined wet film thickness (WFT) on a flat surface where it is dried until a flexible film is obtained. Depending on the equipment and production
⁎
size, the film can be cast into petri dishes (Bansal et al., 2013; ElSetouhy and El-Malak, 2010; Murata et al., 2010) or onto intermediate liners (Preis et al., 2014c; Woertz and Kleinebudde, 2015), which are generally removed before further handling or packaging. For casting the film onto intermediate liners, different continuous and non-continuous manufacturing machines are available. For small-scale experiments or batch productions in a hospital-pharmacy (Visser et al., 2017), a film applicator can be used which consists of a temperature controlled vacuum plate where the intermediate liner is fixed. Afterwards, a coating knife allows casting the film solution with a defined WFT and controlled velocity onto the intermediate liner. In comparison to these small-scale batches, industry produces orodispersible films and patches with larger machines, which usually work continuously (Hoffmann et al., 2011). The intermediate liner is coiled up onto big jumbo rolls. After the film coating, these rolls are cut into smaller daughter rolls and then cut into the final size and packaged (Hoffmann et al., 2011). Due to the continuous manufacturing, ODFs have to dry in a short time period by passing an oven or heating plates. Therefore, a controlled heating of the wet film is required to evaporate the solvents. There are several film coating machines available on the market, which differ among other things in their heating process. Some machines are
Corresponding author E-mail addresses: [email protected] (Y. Thabet), [email protected] (J. Breitkreutz).
https://doi.org/10.1016/j.ijpharm.2017.11.021 Received 15 September 2017; Received in revised form 9 November 2017; Accepted 10 November 2017 Available online 13 November 2017 0378-5173/ © 2017 Elsevier B.V. All rights reserved.
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
equipped with warm air circulation to dry the film, whereas other machines provide electrical heating elements (Optimags, 2017a). Depending on the film formulation, high drying temperature could lead to mechanical instability due to the higher solvent evaporation or degradation of the API caused by high temperature exposure. Especially for poorly water-soluble drugs organic solvents might be necessary to successfully incorporate the API into the ODF. On the one hand the low boiling point of these solvents might lead to problems when drying with high temperatures, such as the formation of air bubbles, which might affect content uniformity issues or degrade the mechanical stability of the film. On the other hand, higher drying temperatures could be advantageous regarding the content of residual solvents (Ph.Eur., 2016). According to the European Pharmacopeia, Class 3 substances may be regarded as less toxic and of lower risk to human health. Class 3 includes common used organic solvents like ethanol or acetone, which may be necessary to dissolve the API. The monograph describes that for Class 3 substances it is considered that amounts of these solvents of 50 mg/day (corresponding to 5000 ppm) are acceptable without justification. Taking into account that about 40% of the drugs with market approval are poorly water-soluble (Kalepu and Nekkanti, 2015), the use of organic solvents during formulation development will be indispensable. Therefore, manufacturing conditions – especially the selection of drying conditions for ODFs – have to be evaluated critically. The ICH Q8 (R2) guideline on Pharmaceutical development (ICH, 2009) describes the advantages of process understandings and underlines the importance of the knowledge of critical process parameters (CPP) and critical quality attributes (CQA). Furthermore, before the pharmaceutical development, a quality target product profile (QTPP) should be created. A QTPP would contain information about the way of delivery (for ODFs the oral delivery because the ODF disintegrates rapidly within the oral cavity and disintegrated particles are swallowed), quality characteristics, safety and efficacy of the dosage form (e.g. high API stability, immediate release kinetic, rapid disintegration to ensure gastrointestinal API absorption, water protected sealed sachets for stability issues). Fig. 1 shows an Ishikawa diagram according to the ICH Q8 guideline on the pharmaceutical development of an orodispersible film. Beside the characteristics of a casting solution (e.g. the viscosity), the production mode plays an important role in product development. During continuous production, the coat of an appropriate film might be insufficient or even not feasible if the viscosity of the solution is too low. Depending on the characteristics of the intermediate liner, contact angles and surface tensions of the casting solutions onto the liner may also have an effect on the resulting film. These effects often correlate with the velocity of the intermediate liner. When the liner moves too
slowly, the viscosity is too low and the contact angle on the surface of the intermediate liner too high, no smooth film will be casted. Even defects of the film might occur. These challenges often do not appear during the non-continuous production. Therefore, a simple scale-up from bench side to continuous production is often impossible. This leads to big challenges in the formulation development and manufacturing at industry scale because previous experiments on bench-side production scale are hardly transferrable to the continuous processes. To our best knowledge, the present paper is the first one dealing with an ODF scale-up process from a lab-scale discontinuous to a pilotscale continuous process. At first, a pilot-scale continuous manufacturing machine is analysed regarding the properties of the equipment and the resulting effects on the orodispersible films. Therefore, viscosity ranges for casting solutions should be defined and effects of different drying temperatures on films casted out of watery-organic casting solutions investigated. In a second step, an ODF containing hydrochlorothiazide (HCT) as example for a Biopharmaceutics Classification System (BCS) Class 4 drug (low solubility, low permeability) (Pires et al., 2011) should be produced with a non-continuous coating bench and afterwards manufactured with a pilot- scale continuous manufacturing machine. The most critical process parameters should be identified during the investigation of the continuous production process. Due to the limited water solubility of HCT acetone is required to dissolve the API in the casting solution. The resulting films were compared regarding their critical quality attributes (e.g. mechanical properties, their content uniformity and residual solvents) to investigate the differences within the manufacturing processes and the effects on the resulting films. 2. Materials and methods 2.1. Materials As film formers hydroxypropylmethylcellulose (HPMC, Pharmacoat 606, Shin-Etsu), and hydroxypropylcellulose (HPC, Klucel, JXF Ashland) were used. Glycerol 85% (Caelo Germany) function as a plasticizer. Distilled water was freshly produced before use. Acetone which was used as a solvent for the film forming was of analytical grade and obtained from VWR international, Germany. Hydrochlorothiazide (HCT) was provided by Boehringer Ingelheim. 2.2. Continuous production of orodispersible films Orodispersible films were continuously produced with a coating machine TGM-K-1.4 (Optimags, Dr. Zimmermann GmbH, Germany)
Fig. 1. Ishikawa diagram of the formulation development of an orodispersible film.
286
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
Fig. 2. Schematic view of the coating machine for continuous film production.
mixtures of acetone/water (50/50) or ethanol/water (50/50). Films were cast at a WFT of 500 μm with an intermediate liner velocity of 125 mm/min. This results in a drying time of 6.5 min. Each film was dried at three different temperature settings: 60/80 °C, 80/100 °C and 100/120 °C (first temperature represents the temperature setting for the first heating element, the latter for the second heating element). Afterwards, the cast films were characterized regarding their mechanical properties and residual solvents. 2.3. Scale up from non-continuous to continuous production exemplified by an orodispersible film containing hydrochlorothiazide Fig. 3. Schematic view of the non-continuous production of HCT films.
Hydrochlorothiazide films consisted of 15% HPC or 15% HPMC, 5.2% HCT and a mixture of acetone/water (50/50).
(Fig. 2). For the coating process an intermediate liner (PPQ 76677, 100 μm silicone coated, Huhtamaki, Finland) is fixed to the machine. The coating technique is a patented coating technique by Optimags (Optimags, 2017b) and the knife (coating width 12 cm) allows the adjustment of the wet film thickness (WFT) of the film. A pump transports the polymer solution through the coating knife onto the intermediate liner. The speed of the intermediate liner can be controlled. The wet film is conveyed through an 80 mm long drying channel, which owns two heating elements. Each heating element can be independently heated up to 120 °C. After the drying process, the film is immediately coiled up onto a jumbo roll. This roll can be cut in daughter rolls and the film can be stamped to the desired size and packed afterwards. In this study, all samples have been cut both, from the centre of the laminate and the outer part, except one centimetre at the edges of the laminate to exclude surface tension and viscoelastic effects.
2.3.1. Non-continuous production Fig. 3 shows the schematic view of a non-continuous production for hydrochlorothiazide films. The casting was conducted on the coating bench/ film applicator Coatmaster 510 (Erichsen, Hemer, Germany), which is equipped with a temperature controlled vacuum plate. HCT films were casted with a velocity of 360 mm/min and a WFT of 500 μm at 20, 50 and 70 °C temperature settings for the vacuum plate. 2.3.2. Continuous production The HCT solutions were casted with a WFT of 500 μm and a velocity of 159 mm/min at three different temperature settings: 60/80 °C, 80/ 100 °C and 100/120 °C. The resulting drying time is 5 min. 2.4. Characterization of the polymer solutions
2.2.1. Characterization of the continuous coating bench The velocity was determined by marking the intermediate liner and measuring the covered distance over a certain time period. Furthermore, the heating zones of the machine were investigated in depth. Therefore, the plate temperatures were measured at 9 points of each heating element at various temperature settings (temperatures from 40 °C to 120 °C) using the LaserSight® thermometer (Optris, Berlin, Germany) equipped with an adapter for metallic surfaces. The percentage deviation of the adjusted value was calculated for each temperature setting and evaluated concerning the position of the measured temperature. To define a viscosity threshold value several polymer solutions ranging from 12–22% solid mass HPMC were cast utilizing the coating machine and evaluated regarding their viscosity and the possibility to cast the solution onto the intermediate liner.
2.4.1. Viscosity of the film solutions Viscosity measurements were performed with a Kinexus rotational viscosimeter (Malvern, UK) utilizing a cone-plate setting (1°) at 25 °C and a shear rate of 6 s−1. The shear rate was chosen according the predominant shear rates on the film coating bench (calculated for applied settings). Each sample was measured three times (each time with 60 viscosity values) and the arithmetic mean and standard deviation (SD) was calculated. 2.5. Characterization of the ODFs 2.5.1. Mass and thickness For characterizing the mass and thickness of the films, ODFs were cut into a sample size of 2 × 3 cm (6 cm2). The thickness was analysed with a micrometer screw (series 331, Mitutoyo, Kawasaki, Japan). Films were weighted with a balance (CP 224 S Sartorius, Göttingen, Germany). Each measurement was performed 6 times. The arithmetic mean and standard deviation were calculated. Samples were measured at the same day of production.
2.2.2. Impact of different drying temperature settings on ODFs based on casting solutions containing different organic solvents The impact of different drying temperatures on the film properties was investigated by casting a 15% HPC film based on water and 287
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
was set to 0.7 ml/min and the HCT was detected at a wavelength of 220 nm. The injection volume was 20 μl. To determine the HCT content, one ODF (sample size 6 cm2) was dissolved in 10.0 mL of the mobile phase and filtered through a 0.45 μm cellulose acetate filter. For the evaluation of the content uniformity 10 film samples were measured. The arithmetic mean and standard deviation were calculated. Samples were measured within one week of production. During the meantime, samples were stored at 25 °C and approx. 40% RH.
2.5.2. Morphological properties Morphological properties of the films were investigated by scanning electron microscopy (SEM) utilizing the G2 Pro microscope (Phenom, Netherlands). ODFs containing HCT were also analysed by polarized light microscopy (Leica Microsystems, Germany) to visualize potential recrystallization processes. Furthermore, a visual inspection of the films was performed. Samples were measured within one week after production. During the meantime, samples were stored at 25 °C and approx. 40% relative humidity (RH).
2.5.8. Disintegration time The disintegration time of the ODFs was determined with a conventional disintegration tester adapted for film formulations (Pharma Test Apparatebau, Hainburg, Germany) (Preis et al., 2014a). Each film (6 cm2) was measured 6 times. The arithmetic mean and standard deviation was calculated. Samples were measured within one week of production. During the meantime, samples were stored at 25 °C and approx. 40% RH.
2.5.3. Mechanical properties The mechanical properties were investigated with a texture analyser (Ta-XTplus, Stable Microsystems, UK) according to Preis et al. (2014b). For the evaluation of the results, the tensile strength and the elongation to break were calculated based on 6 samples. The arithmetic mean and standard deviation were calculated. Samples (6 cm2) were measured at the same day of production. 2.5.4. Loss on drying by infrared light The moisture content of the ODFs was determined by measuring the loss of drying utilizing the heat balance MA 45 Moisture Analyzer (Sartorius, Göttingen, Germany). Temperature was set to 105 °C. The end-point determination was performed in the automatic mode. 3 samples (3 × 3 cm) per batch were measured at the same day of production and the mean was calculated.
3. Results and discussion 3.1. Characterization of the continuous coating bench The setup of the coating machine was investigated to get an idea what characteristics a film solution has to provide to be coated with the machine and to analyse the reproducibility of the film coating. The velocity of the intermediate liner was measured to ensure a reproducible coating process. The measurements show a linear but not proportional correlation between settings and measured velocity with an offset of 20 a.u. (R = 0.999). The setting of 300 leads to a liner velocity of 89 mm/min, whereas the adjustment to 600 leads not to the doubled velocity but to a speed of 198 mm/min (Table 1). Therefore a reproducible coating with a precise adjustment of the liner velocity is feasible. When working with organic solvents, heat exposure on the film has a crucial impact on the quality of the resulting product. The measured temperatures at the heating elements reveal a scattered heating distribution on the heating elements – probably due to the heating mechanism. The elements are heated with electrical wires underneath the metal plates. This results in an inhomogeneous heating distribution. The mean temperature in the middle of the first heating element is 90% of the adjusted value. The temperature decreases in every direction (a mean of 77% of the adjusted value in the upper right corner of the heating element and a mean of 48% in the lower right corner). The heating distribution of the second heating element shows a similar pattern with 90% of the set temperature in the middle of the heating element and only around 60% in the corners of the heating element. This has to be kept in mind while choosing the temperature settings for a film coating. A temperature setting of e.g. 120 °C lead to temperatures of 40 °C in the corner of the heating elements. No uniform drying can be guaranteed, which might be important for e.g. residual solvent analysis. Another point to consider is the viscosity of the film solutions,
2.5.5. Dynamic vapour sorption Drug free films (3 × 3 cm), which were produced to investigate the effect of the drying temperatures, were also tested with a dynamic vapour sorption system SPS11 (Projekt-Messtechnik, ProUmid, Ulm, Germany). Here, the water absorbency after different production conditions can be evaluated. At first the samples were dried down to 0% relative humidity (RH) conditions. The RH was increased with a step size of 10% to a RH of 90%. Every stage was kept until equilibrium of weight was reached for all samples. Afterwards, desorption behaviour was examined by decreasing the RH to 0% again using the same step size. Samples were measured at the same day of production at 25 °C. 2.5.6. Residual solvents The content of residual solvents was measured directly after production when the film has been dried by gas chromatography (GC 6890, Agilent USA) using headspace injection system 7694 (Agilent, USA). The analysis was performed with a Zebron ZB 624 (30 m × 0.32 mm, 1.8 μm film thickness, Phenomenex, USA) column at a helium flow rate of 3 ml/min. The oven was heated to 30 °C for 4 min and then heated up to 150 °C with a heating rate of 15 °C/min; the temperature was hold for 2 min. Residual solvents were detected with a flame ionization detector at 230 °C. The method was validated according to the ICH guideline Q2 Validation of analytical methods (ICH, 2015). To detect even small amounts of solvent, 5 samples (6 cm2 each) were dissolved in water and inserted into the headspace injector. The headspace vials were heated up to 95 °C and equilibrated 15 min before injecting into the gas chromatograph. Samples were measured 24 h after production to ensure complete drying. During the meantime, samples were stored at 25 °C and approx. 40% RH.
Table 1 Machine settings for liner velocity and the resulting speed of the intermediate liner (n = 2).
2.5.7. Hydrochlorothiazide assay The content of hydrochlorothiazide was determined by a high performance liquid chromatography (HPLC) method, which was validated according to the ICH guideline Q2 (ICH, 2015). An Elite La Chrome System (Hitachi-VWR, Darmstadt, Germany) was used, which was equipped with an autosampler L-2200, oven L-2300 and an UV Detector L-2400. As column a 240 × 4 mm Nucleosil RP-18 product was used with a pore size of 5 μm (Macherey-Nagel, Düren, Germany) and operated at 30 °C oven temperature. As mobile phase a mixture of acetonitrile/ phosphate buffer pH 2.2 (50/50) was chosen. The flow rate 288
machine settings [a.u.]
Velocity of the intermediate liner [mm/min]
100 200 250 300 350 400 450 500 600 700 999
17 50 73 89 105 125 144 159 198 235 339
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
Table 2 Dynamic viscosities of HPMC solutions with varying polymer content measured at 6 s−1 and 25 °C, n = 3, mean ± SD. HPMC amount [%]
Dynamic viscosity [Pa*s]
12 14 15 16 18 20 22
0.47 0.90 1.04 1.40 2.30 3.92 5.66
± ± ± ± ± ± ±
Table 3 Weights of drug free films (sample size 6 cm2), n = 6, mean ± RSD.
0.01 0.10 0.01 0.02 0.22 0.03 0.07
which has to exceed a certain threshold value at defined basic settings for intermediate liner velocity and WFT. Below this value, the casting solution would not be casted onto the intermediate liner but flow down at the opposite site of the coating knife. For viscosity analysis the predominant shear rate within the coating knife is of essential importance. The shear rate is dependent on the coating velocity. Preliminary experiments showed that coating velocities of 125–159 mm/min lead to optimal films. Therefore, the shear rate was calculated with a velocity of 159 mm/min. For most casted films a WFT of 400–500 μm was chosen due to the drying time and the drug load of the film. This leads to a calculated shear rate of 6 s−1. Viscosity measurements were all performed at this shear rate to imitate the conditions inside the coating knife. Table 2 displays the obtained viscosities of polymeric solutions for the determination of viscosity ranges. All HPMC solutions were tried to be casted at a WFT of 500 μm with a velocity of 159 mm/min. Solutions of 15% HPMC or higher were able to be casted onto the intermediate liner. Lower polymer content lead to a leaking on the front side of the coating knife, so that the polymer solution is not transported onto the intermediate liner. A 15% HPMC solution reveals a dynamic viscosity of 1 Pa*s. Therefore, under usual manufacturing conditions, polymer solutions should have at least a dynamic viscosity of 1 Pa*s. In addition it was observed that higher viscosities lead to reduced spreading of the coating solution onto the intermediate liner, which effects the API content of the film because the API is distributed inside the polymer solution and dosed via the area of the film. A high spreading of the polymer solution leads to a lower wet film thickness and a lower API content per cm2. These effects are not that pronounced during the non-continuous production because the casted film area is larger so that edge effects are not that preponderated. In contrast to the non-continuous film production, the continuous production always requires temperature zones to dry the film in the short time period before the film is coiled up. Especially when working with organic solvents, the drying temperatures might affect the mechanical properties of the films and the content of residual solvents. Therefore, drug-free HPC films were investigated after production dried at different drying conditions. With dynamic viscosities of 3.11 ± 0.01 Pa*s (aqueous solution), 4.50 ± 0.84 Pa*s (acetone/ water) and 3.69 ± 0.03 Pa*s (ethanol/water) all casting film solutions can be casted onto the intermediate liner. All production conditions from 60/80 °C to 100/120 °C led to flexible, transparent films without recrystallization or precipitation of any component. Due to the solvent evaporation, higher amounts of air bubbles can be observed inside films dried at high temperatures, especially at the centre of the film, due to the heat distribution of the heating elements. Therefore, only the weight and not the film thickness was chosen as representative parameter because the measurements with the micrometre screw led to high variations due to the air bubbles. Table 3 displays the results for the weight measurements. Films casted of an acetone/ water mixture weigh less than films casted out of a watery solution. Ethanolic films showed the highest weights but also revealed the highest relative standard deviation (RSD). No uniform drying of ethanolic films was achieved. Fig. 4 shows the mechanical characterization of the films based on casting solutions with different solvent mixtures. Different drying
HPC films
Weight [mg/6 cm2]
Relative standard deviation [%]
Acetone 60/80 °C Acetone 80/100 °C Acetone 100/120 °C Ethanol 60/80 °C Ethanol 80/ 100 °C Ethanol 100/120 °C Water 60/80 °C Water 80/100 °C Water 100/120 °C
20.9 18.5 17.2 29.0 27.7 27.1 23.1 23.4 22.6
± 2.1 ± 0.7 ± 2.1 ± 5.2 ± 7.1 ± 7.4 ± 3.6 ± 3.8 ± 2.3
Fig. 4. Puncture Strength of drug free films casted out of organic solvent mixtures, dried at various drying temperatures (60/80 °C, 80/100 °C, 100/120 °C), n = 6, mean ± SD.
conditions have no significant impact on HPC films cast out of an ethanol/ water mixture, whereas higher drying temperatures lead to a decrease of puncture strength for films casted out of acetone/ water mixtures. This can be explained with the lower boiling point of acetone, which leads to more solvent evaporation and less flexibility and strength at higher drying temperatures. Films based on water show a trend to higher puncture strengths with increasing drying temperatures. For the highest selected drying temperatures, ethanolic films revealed the highest elongation to break (15 ± 4%) followed by aqueous films with an elongation of 10 ± 2%. Films casted out of an acetone/ water mixture showed the lowest elongation to break with 8 ± 3%. The results of the dynamic vapour sorption (Fig. 5) showed no differences on the absorption capacity of different drug free films. Once dried, water sorption remains constant independently of the solvent used for film preparation. Only in the beginning of the measurement, when the relative humidity is at 0%, films based on an acetone/water
Fig. 5. Dynamic vapour sorption of drug-free ODFs casted from film solutions containing acetone, ethanol and water.
289
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
temperatures. A threshold value of 1 Pa*s could be defined for the dynamic viscosity of the casting solutions under common production conditions. Less viscous solutions were not able to be cast onto the intermediate liner but flew down at the opposite site of the coating knife. Drug-free films dried at different drying temperatures show differences in the content of organic solvents and small differences in their mechanical properties. The water absorption capacity is not affected by the selected drying temperatures. Drug-free films casted out of an acetonic solution reveal more air bubbles than films of ethanolic or aqueous casting solutions. The higher the selected drying temperature, the higher the air bubble formation during the drying process. This could lead to variations in content uniformity for API containing films. 3.2. Scale up for an HCT ODF from non-continuous to continuous production
Fig. 6. Loss of drying of drug-free films. Relative humidity was set to 0%. The diagram shows the loss of mass in % over the time.
Previously described results were considered for the scale-up of an orodispersible film containing HCT. The HCT castings solutions were first casted in a non-continuous process simulating the continuous production as far as possible by setting the drying temperatures to 50 °C or 70 °C. The same polymer solutions were subsequently casted utilizing the continuous film coater. A comparison of the resulting films show the differences and the critical process parameters during the production process. ODFs containing HPC as film former led in all cases to transparent and flexible films, irrespectively of the selected drying temperatures. In contrast ODFs made from HPMC resulted in transparent films directly after production, but precipitation was observed after approximately 12 h for the continuous as well as the non- continuous production. Polarised light microscopy reveal no crystals within the film. This leads to the assumption that HPMC precipitates in an amorphous state forming irregular “clouds”. Elevated drying temperatures decreased the amount and sizes of the formed precipitates. The higher the drying temperatures, the lower was the degree of precipitation. In the continuous production mode less precipitation was observed in comparison to the films casted with the Erichsen coater. At drying conditions of 100/120 °C no precipitation was observed at all, but these drying conditions are applicable for the continuous process only. Fig. 7 shows the results of the mechanical properties of HCT films consisting of different polymers, dried at different drying temperatures in either continuous or non-continuous production mode. A distinct difference between HPMC and HPC films can be observed, especially for the non-continuous production. HPC films reveal a lower puncture strength then HPMC films. In contrast to these results, they have a
mixture show lower mass loss compared to films casted out of watery or ethanol/water solutions (Fig. 6). Results of the moisture analysis reveal that with higher drying temperatures less moisture (defined as residual ethanol or acetone in addition to residual water) remains inside the ODFs directly after production (for water based films: 5.4% at 60/80 °C and 3.3% at 100/ 120 °C). Furthermore, at lowest drying temperatures, moisture contents of water based films are the highest (5.4%) whereas the water content decreases when organic solvents were incorporated (4.5% ethanol/ water mixtures and 4.0% acetone/water mixtures). These effects can be explained with the lower boiling point of ethanol and acetone which enables them to evaporate faster than water. Nevertheless, all casted films revealed adequate mechanical properties as they could be coiled up immediately after production. Further handling was feasible without breaking the films. Table 4 displays the results for the gas chromatographic analysis of residual ethanol or acetone of the films directly after production. For films based on an ethanol/ water mixture, the amount of ethanol decreases with increasing drying temperatures. A single film piece contained 2773 ppm ethanol when dried at 60/80 °C, whereas drying at 100/120 °C led to a lower ethanol content of 191 ppm ethanol per single dose. For the films based on an acetone/water solution, no acetone peak could be observed in the chromatogram directly after production. Therefore the content had to be below the detection limit of the method (vial concentration of 0.3 ppm). The low boiling point of acetone led to almost complete evaporation of the solvent during all production conditions. The characterization of the continuous film coating machine showed that coating process is reproducible because the heating and the velocity of the release liner can be precisely adjusted. But the drying of the films is inconsistent during the drying process due to the irregular heat distribution of the heating elements. This leads to centred air bubble formation when working with organic solvents at higher drying
Table 4 Results from GC analysis of residual solvents, mean, n = 2. Solvents basis
Drying condition
Content of residual solvents/ single dose [ppm]
ethanol/water (50/ 50)
60/80 °C 80/100 °C 100/120 °C
2773 910 191
acetone/water (50/ 50)
60/80 °C 80/100 °C 100/120 °C
n.d.a n.d.a n.d.a
a
Fig. 7. Puncture strengths of HCT films made from HPC or HPMC solutions at different temperatures in bench scale and continuous manufacturing mode, n = 6; mean ± SD.
n.d. = acetone content below detection limit.
290
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
Fig. 8. Residual acetone of HCT films based on HPC, dashed line represents the limit for residual solvents (Class 3) according to the Ph. Eur. n = 2, mean. Fig. 9. Content HCT per single dose measured by HPLC, n = 10, mean ± SD.
higher puncture strength with increasing temperatures continuously produced. Therefore, HPC films reveal different mechanical properties in dependency of their production conditions. This effect is more pronounced for the HPC than HPMC films, which exhibit no major changes in puncture strength when varying the drying condition or the mode of the production process. The results of the gas chromatographic analysis to determine the content of residual solvents of HPC films are displayed in Fig. 8. With increasing drying temperatures, the amount of residual acetone decreases. Therefore ODFs produced with the continuous coating machine contain less acetone than films produced in a noncontinuous production mode. This is most probably due to the higher drying temperatures. In comparison to the results obtained for the drugfree HPC films, HCT films reveal a higher acetone content. The presence of an active pharmaceutical ingredient clearly affects the content of residual solvents. According to the Ph.Eur. for adults an amount of less than 5000 ppm per day is acceptable without justification (Ph.Eur., 2016). However, for paediatric purposes lower concentrations may be appropriate. All films prepared using the continuous machine reveal less than 2100 ppm acetone per single dose directly after production and therefore fulfil the requirements regarding the content of residual solvents, whereas the non-continuous produced films show a higher content of residual solvents despite they were analysed 24 h after production to ensure complete dryness of the films. These results illustrate that when working with organic solvents, the scale-up from a noncontinuous to a continuous process is not feasible. Resulting concentrations of residual solvents are not predictable. No conclusions can be drawn from a bench-scale to a continuous process; even if the process settings are known. Different APIs may affect the content of residual solvents. Therefore, for each formulation and every change in process parameters the content has to be determined. The content uniformity of HCT (Fig. 9) in HPMC films is not affected by different drying temperatures neither in the continuous nor in the non-continuous production mode. HPC films generally reveal a lower content of HCT than HPMC films per area of the predefined film piece, especially with higher drying temperatures in the non-continuous production. However, the content uniformity is not affected by this phenomenon (relative SD of 3–6%). ODFs consisting of HPC produced within the continuous production, contain less HCT per single dose than HPMC films (7 mg with HPC as polymer or 13 mg with HPMC as polymer), but are not affected by the different drying temperatures. This confirms that the viscosity of the film solution, in combination with the chosen coating velocity plays an important role for the continuous film casting. A high molecular HPC was chosen as film forming polymer to reduce the spreading effects onto the release liner during the coating process. This leads to a high casting viscosity of 20.4 Pa*s in contrast to the HPMC solutions utilizing a low molecular HPMC, which reveal a viscosity of 2.1 ± 0.08 Pa*s near to the determined threshold value for film casting.
All continuously produced HCT films disintegrate within less than 180 s (max. 14 ± 1 s) and therefore fulfil the requirements of orodispersible dosage forms according to Ph.Eur. (2017). Only the HCT films consisting of HPC dried at 20 °C needed with 67 ± 11 s more than 30 s for disintegration. The comparison of the HCT films in a continuous and non-continuous production mode could show high differences for the content of residual solvents. Continuously produced ODFs reveal a distinct lower amount of acetone than the bench- scale products. This can be explained with the higher drying temperatures. Lab-scale films produced at 70 °C contain a similar amount of residual acetone than the continuously produced ODFs. The production mode also affects the API content of HPC films. Where the HPMC films contain almost the same amount for every production setting, the HCT content is decreasing for HPC films with higher drying temperatures. This can be explained with the spreading of the film solution on the intermediate liner. The higher the selected drying temperature, the less time for spreading and the higher the layer thickness during drying. The mechanical properties are not highly affected by the selection of the production mode. Only the HPC films show differences in puncture strength during bench- scale production. In conclusion, the most critical process parameters for an up-scaling are the viscosity of the coating solution and the selected drying temperatures. 4. Conclusion This study shows that a transfer from a lab-scale to a continuous pilot-scale production is not easy to be performed. Depending on the film-forming polymer more or less differences of film properties due to the casting process can be observed. HPMC films are less affected by the production mode, but display temperature dependent precipitations. HPC films show higher differences depending on the production mode and conditions. Lab-scale production leads to higher drug content but also to higher content of residual solvents. The mechanical strength (expressed as puncture strength) is improved by the continuous production mode. Different requirements on the film solutions also underline the outcome that a transfer from bench-scale to continuous production cannot be easily performed. In conclusion, when transferring film manufacturing from non-continuous to continuous mode, especially the viscosity of the casting solution and the drying temperatures are critical process parameters which cannot be simulated during a lab-scale production on a bench. These parameters have to be selected individually for each product when transferring the process to a pilot-scale or production-scale continuous process. In our opinion, it is even questionable weather discontinuous lab-scale experiments should be conducted if the target should be a production-scale or whether 291
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
dissolving films for oral dosage from natural polysaccharides. Materials 3, 4291–4299. Optimags, 2017a. Dr. Zimmermann GmbH, TDS ODV Development Equipment [Accessed 06 September 2017]. http://www.optimags.de/development_equipment.php. Optimags, 2017b. Optimags Coating Machines [Accessed 07 November 2017]. http:// www.optimags.de/advantages_of_optimags.php. Ph.Eur, 2014. European Pharmacopoeia Comission, Oromucosal Preparations, European Pharmacopoeia. European Directorate for the Quality of Medicines & Healthcare (EDQM), Strasbourg, France. Ph.Eur, 2016. European Pharmacopoeia Commission, Residual Solvents, European Pharmacopoeia. European Directorate for the Quality of Medicines & Healthcare (EDQM), Strasbourg, France. Ph.Eur, 2017. European Pharmacopoeia Comission, Orodispersible Tablets, European Pharmacopoeia. European Directorate for the Quality of Medicines & Healthcare (EDQM), Strasbourg, France. Pires, M.A.S., Souza dos Santos, R.A., Sinisterra, R.D., 2011. Pharmaceutical composition of hydrochlorothiazide: β-cyclo-dextrin: preparation by three different methods, physico-chemical characterization and in vivo diuretic activity evaluation. Molecules 16, 4482–4499. Preis, M., 2015. Orally disintegrating films and mini-tablets—innovative dosage forms of choice for pediatric use. AAPS PharmSciTech 16, 234–241. Preis, M., Gronkowsky, D., Grytzan, D., Breitkreutz, J., 2014a. Comparative study on novel test systems to determine disintegration time of orodispersible films. J. Pharm. Pharmacol. 66, 1102–1111. Preis, M., Knop, K., Breitkreutz, J., 2014b. Mechanical strength test for orodispersible and buccal films. Int. J. Pharm. 461, 22–29. Preis, M., Woertz, C., Schneider, K., Kukawka, J., Broscheit, J., Roewer, N., Breitkreutz, J., 2014c. Design and evaluation of bilayered buccal film preparations for local administration of lidocaine hydrochloride. Eur. J. Pharm. Biopharm. 86, 552–561. Visser, J.C., Woerdenbag, H.J., Hanff, L.M., Frijlink, H.W., 2017. Personalized medicine in pediatrics: the clinical potential of orodispersible films. AAPS PharmSciTech 18, 267–272. Woertz, C., Kleinebudde, P., 2015. Development of orodispersible polymer films containing poorly water soluble active pharmaceutical ingredients with focus on different drug loadings and storage stability. Int. J. Pharm. 493, 134–145.
continuous pilot-scale equipment should be used from the beginning. Funding sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References Bansal, S., Bansal, M., Garg, G., 2013. Formulation and evaluation of fast dissolving film of an antihypertensive drug. Int. J. Pharm. Chem. Biol. Sci. 3, 1097–1108. Buanz, A.B., Belaunde, C.C., Soutari, N., Tuleu, C., Gul, M.O., Gaisford, S., 2015. Ink-jet printing versus solvent casting to prepare oral films: effect on mechanical properties and physical stability. Int. J. Pharm. 494, 611–618. Cilurzo, F., Cupone, I., Minghetti, P., Selmin, F., Montanari, L., 2008. Fast dissolving films made of maltodextrins. Eur. J. Pharm. Biopharm. 70, 895–900. Dixit, R., Puthli, S., 2009. Oral strip technology: overview and future potential. J. Control. Release 139, 94–107. El-Setouhy, D.A., El-Malak, N.S.A., 2010. Formulation of a novel tianeptine sodium orodispersible film. AAPS PharmSciTech 11, 1018–1025. Hoffmann, E.M., Breitenbach, A., Breitkreutz, J., 2011. Advances in orodispersible films for drug delivery. Expert Opin. Drug Deliv. 8, 299–316. ICH, 2009. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH), Pharmaceutical Development, Q8 (R2). ICH, 2015. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH), Validation of Analytical Procedures: Text and Methodology Q2 (R1). Jani, R., Patel, D., 2015. Hot melt extrusion: an industrially feasible approach for casting orodispersible film. Asian J. Pharm. Sci. 10, 292–305. Kalepu, S., Nekkanti, V., 2015. Insoluble drug delivery strategies: review of recent advances and business prospects. Acta Pharm. Sin. B 5, 442–453. Murata, Y., Isobe, T., Kofuji, K., Nishida, N., Kamaguchi, R., 2010. Preparation of fast
292
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Research Paper
Orodispersible films: Product transfer from lab-scale to continuous manufacturing Yasmin Thabet, Joerg Breitkreutz
T
⁎
Heinrich Heine University Duesseldorf, Universitaetsstrasse 1, 40225, Duesseldorf, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Orodispersible films Oral thin strips Film preparations Continuous manufacturing Hydrochlorothiazide Scaling-up
Orodispersible films have been described as new beneficial dosage forms for special patient populations. Due to various production settings, different requirements on film formulations are required for non- continuous and continuous manufacturing. In this study, a continuous coating machine was qualified in regards of the process conditions for film compositions and their effects on the formed films. To investigate differences between both manufacturing processes, various film formulations of hydrochlorothiazide and hydroxypropylcellulose (HPC) or hydroxypropylmethycellulose (HPMC) as film formers were produced and the resulting films were characterized. The qualification of the continuously operating coating machine reveals no uniform heat distribution during drying. Coating solutions for continuous manufacturing should provide at least a dynamic viscosity of 1 Pa*s (wet film thickness of 500 μm, velocity of 15.9 cm/min). HPC films contain higher residuals of ethanol or acetone in bench-scale than in continuous production mode. Continuous production lead to lower drug content of the films. All continuously produced films disintegrate within less than 30 s. There are observed significant effects of the production process on the film characteristics. When transferring film manufacturing from lab-scale to continuous mode, film compositions, processing conditions and suitable characterization methods have to be carefully selected and adopted.
1. Introduction Orodispersible films (ODFs) are defined as “single- or multilayer sheets of suitable materials, to be placed in the mouth where they disperse rapidly” (Ph.Eur., 2014). This dosage form provides the advantage of fully flexible dosing for several patient populations by cutting the film into the desired size. Furthermore, the film does not have to be swallowed like a conventional tablet, but only placed onto the tongue where it disperses rapidly into smaller pieces, which could then easily be swallowed. Therefore, paediatrics or geriatrics are potential target populations, which often suffer from swallowing difficulties (Hoffmann et al., 2011; Preis, 2015). There are several production techniques for ODFs described in literature. Beside the infrequently used hot melt extrusion process (Jani and Patel, 2015), the most common production method is via solvent casting method (Buanz et al., 2015; Cilurzo et al., 2008). Here, the active pharmaceutical ingredient (API) is dissolved or dispersed in a polymer solution. Additional ingredients like plasticizers, taste masking agents or fillers may be necessary (Dixit and Puthli, 2009). Afterwards, the solution is cast with a defined wet film thickness (WFT) on a flat surface where it is dried until a flexible film is obtained. Depending on the equipment and production
⁎
size, the film can be cast into petri dishes (Bansal et al., 2013; ElSetouhy and El-Malak, 2010; Murata et al., 2010) or onto intermediate liners (Preis et al., 2014c; Woertz and Kleinebudde, 2015), which are generally removed before further handling or packaging. For casting the film onto intermediate liners, different continuous and non-continuous manufacturing machines are available. For small-scale experiments or batch productions in a hospital-pharmacy (Visser et al., 2017), a film applicator can be used which consists of a temperature controlled vacuum plate where the intermediate liner is fixed. Afterwards, a coating knife allows casting the film solution with a defined WFT and controlled velocity onto the intermediate liner. In comparison to these small-scale batches, industry produces orodispersible films and patches with larger machines, which usually work continuously (Hoffmann et al., 2011). The intermediate liner is coiled up onto big jumbo rolls. After the film coating, these rolls are cut into smaller daughter rolls and then cut into the final size and packaged (Hoffmann et al., 2011). Due to the continuous manufacturing, ODFs have to dry in a short time period by passing an oven or heating plates. Therefore, a controlled heating of the wet film is required to evaporate the solvents. There are several film coating machines available on the market, which differ among other things in their heating process. Some machines are
Corresponding author E-mail addresses: [email protected] (Y. Thabet), [email protected] (J. Breitkreutz).
https://doi.org/10.1016/j.ijpharm.2017.11.021 Received 15 September 2017; Received in revised form 9 November 2017; Accepted 10 November 2017 Available online 13 November 2017 0378-5173/ © 2017 Elsevier B.V. All rights reserved.
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
equipped with warm air circulation to dry the film, whereas other machines provide electrical heating elements (Optimags, 2017a). Depending on the film formulation, high drying temperature could lead to mechanical instability due to the higher solvent evaporation or degradation of the API caused by high temperature exposure. Especially for poorly water-soluble drugs organic solvents might be necessary to successfully incorporate the API into the ODF. On the one hand the low boiling point of these solvents might lead to problems when drying with high temperatures, such as the formation of air bubbles, which might affect content uniformity issues or degrade the mechanical stability of the film. On the other hand, higher drying temperatures could be advantageous regarding the content of residual solvents (Ph.Eur., 2016). According to the European Pharmacopeia, Class 3 substances may be regarded as less toxic and of lower risk to human health. Class 3 includes common used organic solvents like ethanol or acetone, which may be necessary to dissolve the API. The monograph describes that for Class 3 substances it is considered that amounts of these solvents of 50 mg/day (corresponding to 5000 ppm) are acceptable without justification. Taking into account that about 40% of the drugs with market approval are poorly water-soluble (Kalepu and Nekkanti, 2015), the use of organic solvents during formulation development will be indispensable. Therefore, manufacturing conditions – especially the selection of drying conditions for ODFs – have to be evaluated critically. The ICH Q8 (R2) guideline on Pharmaceutical development (ICH, 2009) describes the advantages of process understandings and underlines the importance of the knowledge of critical process parameters (CPP) and critical quality attributes (CQA). Furthermore, before the pharmaceutical development, a quality target product profile (QTPP) should be created. A QTPP would contain information about the way of delivery (for ODFs the oral delivery because the ODF disintegrates rapidly within the oral cavity and disintegrated particles are swallowed), quality characteristics, safety and efficacy of the dosage form (e.g. high API stability, immediate release kinetic, rapid disintegration to ensure gastrointestinal API absorption, water protected sealed sachets for stability issues). Fig. 1 shows an Ishikawa diagram according to the ICH Q8 guideline on the pharmaceutical development of an orodispersible film. Beside the characteristics of a casting solution (e.g. the viscosity), the production mode plays an important role in product development. During continuous production, the coat of an appropriate film might be insufficient or even not feasible if the viscosity of the solution is too low. Depending on the characteristics of the intermediate liner, contact angles and surface tensions of the casting solutions onto the liner may also have an effect on the resulting film. These effects often correlate with the velocity of the intermediate liner. When the liner moves too
slowly, the viscosity is too low and the contact angle on the surface of the intermediate liner too high, no smooth film will be casted. Even defects of the film might occur. These challenges often do not appear during the non-continuous production. Therefore, a simple scale-up from bench side to continuous production is often impossible. This leads to big challenges in the formulation development and manufacturing at industry scale because previous experiments on bench-side production scale are hardly transferrable to the continuous processes. To our best knowledge, the present paper is the first one dealing with an ODF scale-up process from a lab-scale discontinuous to a pilotscale continuous process. At first, a pilot-scale continuous manufacturing machine is analysed regarding the properties of the equipment and the resulting effects on the orodispersible films. Therefore, viscosity ranges for casting solutions should be defined and effects of different drying temperatures on films casted out of watery-organic casting solutions investigated. In a second step, an ODF containing hydrochlorothiazide (HCT) as example for a Biopharmaceutics Classification System (BCS) Class 4 drug (low solubility, low permeability) (Pires et al., 2011) should be produced with a non-continuous coating bench and afterwards manufactured with a pilot- scale continuous manufacturing machine. The most critical process parameters should be identified during the investigation of the continuous production process. Due to the limited water solubility of HCT acetone is required to dissolve the API in the casting solution. The resulting films were compared regarding their critical quality attributes (e.g. mechanical properties, their content uniformity and residual solvents) to investigate the differences within the manufacturing processes and the effects on the resulting films. 2. Materials and methods 2.1. Materials As film formers hydroxypropylmethylcellulose (HPMC, Pharmacoat 606, Shin-Etsu), and hydroxypropylcellulose (HPC, Klucel, JXF Ashland) were used. Glycerol 85% (Caelo Germany) function as a plasticizer. Distilled water was freshly produced before use. Acetone which was used as a solvent for the film forming was of analytical grade and obtained from VWR international, Germany. Hydrochlorothiazide (HCT) was provided by Boehringer Ingelheim. 2.2. Continuous production of orodispersible films Orodispersible films were continuously produced with a coating machine TGM-K-1.4 (Optimags, Dr. Zimmermann GmbH, Germany)
Fig. 1. Ishikawa diagram of the formulation development of an orodispersible film.
286
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
Fig. 2. Schematic view of the coating machine for continuous film production.
mixtures of acetone/water (50/50) or ethanol/water (50/50). Films were cast at a WFT of 500 μm with an intermediate liner velocity of 125 mm/min. This results in a drying time of 6.5 min. Each film was dried at three different temperature settings: 60/80 °C, 80/100 °C and 100/120 °C (first temperature represents the temperature setting for the first heating element, the latter for the second heating element). Afterwards, the cast films were characterized regarding their mechanical properties and residual solvents. 2.3. Scale up from non-continuous to continuous production exemplified by an orodispersible film containing hydrochlorothiazide Fig. 3. Schematic view of the non-continuous production of HCT films.
Hydrochlorothiazide films consisted of 15% HPC or 15% HPMC, 5.2% HCT and a mixture of acetone/water (50/50).
(Fig. 2). For the coating process an intermediate liner (PPQ 76677, 100 μm silicone coated, Huhtamaki, Finland) is fixed to the machine. The coating technique is a patented coating technique by Optimags (Optimags, 2017b) and the knife (coating width 12 cm) allows the adjustment of the wet film thickness (WFT) of the film. A pump transports the polymer solution through the coating knife onto the intermediate liner. The speed of the intermediate liner can be controlled. The wet film is conveyed through an 80 mm long drying channel, which owns two heating elements. Each heating element can be independently heated up to 120 °C. After the drying process, the film is immediately coiled up onto a jumbo roll. This roll can be cut in daughter rolls and the film can be stamped to the desired size and packed afterwards. In this study, all samples have been cut both, from the centre of the laminate and the outer part, except one centimetre at the edges of the laminate to exclude surface tension and viscoelastic effects.
2.3.1. Non-continuous production Fig. 3 shows the schematic view of a non-continuous production for hydrochlorothiazide films. The casting was conducted on the coating bench/ film applicator Coatmaster 510 (Erichsen, Hemer, Germany), which is equipped with a temperature controlled vacuum plate. HCT films were casted with a velocity of 360 mm/min and a WFT of 500 μm at 20, 50 and 70 °C temperature settings for the vacuum plate. 2.3.2. Continuous production The HCT solutions were casted with a WFT of 500 μm and a velocity of 159 mm/min at three different temperature settings: 60/80 °C, 80/ 100 °C and 100/120 °C. The resulting drying time is 5 min. 2.4. Characterization of the polymer solutions
2.2.1. Characterization of the continuous coating bench The velocity was determined by marking the intermediate liner and measuring the covered distance over a certain time period. Furthermore, the heating zones of the machine were investigated in depth. Therefore, the plate temperatures were measured at 9 points of each heating element at various temperature settings (temperatures from 40 °C to 120 °C) using the LaserSight® thermometer (Optris, Berlin, Germany) equipped with an adapter for metallic surfaces. The percentage deviation of the adjusted value was calculated for each temperature setting and evaluated concerning the position of the measured temperature. To define a viscosity threshold value several polymer solutions ranging from 12–22% solid mass HPMC were cast utilizing the coating machine and evaluated regarding their viscosity and the possibility to cast the solution onto the intermediate liner.
2.4.1. Viscosity of the film solutions Viscosity measurements were performed with a Kinexus rotational viscosimeter (Malvern, UK) utilizing a cone-plate setting (1°) at 25 °C and a shear rate of 6 s−1. The shear rate was chosen according the predominant shear rates on the film coating bench (calculated for applied settings). Each sample was measured three times (each time with 60 viscosity values) and the arithmetic mean and standard deviation (SD) was calculated. 2.5. Characterization of the ODFs 2.5.1. Mass and thickness For characterizing the mass and thickness of the films, ODFs were cut into a sample size of 2 × 3 cm (6 cm2). The thickness was analysed with a micrometer screw (series 331, Mitutoyo, Kawasaki, Japan). Films were weighted with a balance (CP 224 S Sartorius, Göttingen, Germany). Each measurement was performed 6 times. The arithmetic mean and standard deviation were calculated. Samples were measured at the same day of production.
2.2.2. Impact of different drying temperature settings on ODFs based on casting solutions containing different organic solvents The impact of different drying temperatures on the film properties was investigated by casting a 15% HPC film based on water and 287
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
was set to 0.7 ml/min and the HCT was detected at a wavelength of 220 nm. The injection volume was 20 μl. To determine the HCT content, one ODF (sample size 6 cm2) was dissolved in 10.0 mL of the mobile phase and filtered through a 0.45 μm cellulose acetate filter. For the evaluation of the content uniformity 10 film samples were measured. The arithmetic mean and standard deviation were calculated. Samples were measured within one week of production. During the meantime, samples were stored at 25 °C and approx. 40% RH.
2.5.2. Morphological properties Morphological properties of the films were investigated by scanning electron microscopy (SEM) utilizing the G2 Pro microscope (Phenom, Netherlands). ODFs containing HCT were also analysed by polarized light microscopy (Leica Microsystems, Germany) to visualize potential recrystallization processes. Furthermore, a visual inspection of the films was performed. Samples were measured within one week after production. During the meantime, samples were stored at 25 °C and approx. 40% relative humidity (RH).
2.5.8. Disintegration time The disintegration time of the ODFs was determined with a conventional disintegration tester adapted for film formulations (Pharma Test Apparatebau, Hainburg, Germany) (Preis et al., 2014a). Each film (6 cm2) was measured 6 times. The arithmetic mean and standard deviation was calculated. Samples were measured within one week of production. During the meantime, samples were stored at 25 °C and approx. 40% RH.
2.5.3. Mechanical properties The mechanical properties were investigated with a texture analyser (Ta-XTplus, Stable Microsystems, UK) according to Preis et al. (2014b). For the evaluation of the results, the tensile strength and the elongation to break were calculated based on 6 samples. The arithmetic mean and standard deviation were calculated. Samples (6 cm2) were measured at the same day of production. 2.5.4. Loss on drying by infrared light The moisture content of the ODFs was determined by measuring the loss of drying utilizing the heat balance MA 45 Moisture Analyzer (Sartorius, Göttingen, Germany). Temperature was set to 105 °C. The end-point determination was performed in the automatic mode. 3 samples (3 × 3 cm) per batch were measured at the same day of production and the mean was calculated.
3. Results and discussion 3.1. Characterization of the continuous coating bench The setup of the coating machine was investigated to get an idea what characteristics a film solution has to provide to be coated with the machine and to analyse the reproducibility of the film coating. The velocity of the intermediate liner was measured to ensure a reproducible coating process. The measurements show a linear but not proportional correlation between settings and measured velocity with an offset of 20 a.u. (R = 0.999). The setting of 300 leads to a liner velocity of 89 mm/min, whereas the adjustment to 600 leads not to the doubled velocity but to a speed of 198 mm/min (Table 1). Therefore a reproducible coating with a precise adjustment of the liner velocity is feasible. When working with organic solvents, heat exposure on the film has a crucial impact on the quality of the resulting product. The measured temperatures at the heating elements reveal a scattered heating distribution on the heating elements – probably due to the heating mechanism. The elements are heated with electrical wires underneath the metal plates. This results in an inhomogeneous heating distribution. The mean temperature in the middle of the first heating element is 90% of the adjusted value. The temperature decreases in every direction (a mean of 77% of the adjusted value in the upper right corner of the heating element and a mean of 48% in the lower right corner). The heating distribution of the second heating element shows a similar pattern with 90% of the set temperature in the middle of the heating element and only around 60% in the corners of the heating element. This has to be kept in mind while choosing the temperature settings for a film coating. A temperature setting of e.g. 120 °C lead to temperatures of 40 °C in the corner of the heating elements. No uniform drying can be guaranteed, which might be important for e.g. residual solvent analysis. Another point to consider is the viscosity of the film solutions,
2.5.5. Dynamic vapour sorption Drug free films (3 × 3 cm), which were produced to investigate the effect of the drying temperatures, were also tested with a dynamic vapour sorption system SPS11 (Projekt-Messtechnik, ProUmid, Ulm, Germany). Here, the water absorbency after different production conditions can be evaluated. At first the samples were dried down to 0% relative humidity (RH) conditions. The RH was increased with a step size of 10% to a RH of 90%. Every stage was kept until equilibrium of weight was reached for all samples. Afterwards, desorption behaviour was examined by decreasing the RH to 0% again using the same step size. Samples were measured at the same day of production at 25 °C. 2.5.6. Residual solvents The content of residual solvents was measured directly after production when the film has been dried by gas chromatography (GC 6890, Agilent USA) using headspace injection system 7694 (Agilent, USA). The analysis was performed with a Zebron ZB 624 (30 m × 0.32 mm, 1.8 μm film thickness, Phenomenex, USA) column at a helium flow rate of 3 ml/min. The oven was heated to 30 °C for 4 min and then heated up to 150 °C with a heating rate of 15 °C/min; the temperature was hold for 2 min. Residual solvents were detected with a flame ionization detector at 230 °C. The method was validated according to the ICH guideline Q2 Validation of analytical methods (ICH, 2015). To detect even small amounts of solvent, 5 samples (6 cm2 each) were dissolved in water and inserted into the headspace injector. The headspace vials were heated up to 95 °C and equilibrated 15 min before injecting into the gas chromatograph. Samples were measured 24 h after production to ensure complete drying. During the meantime, samples were stored at 25 °C and approx. 40% RH.
Table 1 Machine settings for liner velocity and the resulting speed of the intermediate liner (n = 2).
2.5.7. Hydrochlorothiazide assay The content of hydrochlorothiazide was determined by a high performance liquid chromatography (HPLC) method, which was validated according to the ICH guideline Q2 (ICH, 2015). An Elite La Chrome System (Hitachi-VWR, Darmstadt, Germany) was used, which was equipped with an autosampler L-2200, oven L-2300 and an UV Detector L-2400. As column a 240 × 4 mm Nucleosil RP-18 product was used with a pore size of 5 μm (Macherey-Nagel, Düren, Germany) and operated at 30 °C oven temperature. As mobile phase a mixture of acetonitrile/ phosphate buffer pH 2.2 (50/50) was chosen. The flow rate 288
machine settings [a.u.]
Velocity of the intermediate liner [mm/min]
100 200 250 300 350 400 450 500 600 700 999
17 50 73 89 105 125 144 159 198 235 339
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
Table 2 Dynamic viscosities of HPMC solutions with varying polymer content measured at 6 s−1 and 25 °C, n = 3, mean ± SD. HPMC amount [%]
Dynamic viscosity [Pa*s]
12 14 15 16 18 20 22
0.47 0.90 1.04 1.40 2.30 3.92 5.66
± ± ± ± ± ± ±
Table 3 Weights of drug free films (sample size 6 cm2), n = 6, mean ± RSD.
0.01 0.10 0.01 0.02 0.22 0.03 0.07
which has to exceed a certain threshold value at defined basic settings for intermediate liner velocity and WFT. Below this value, the casting solution would not be casted onto the intermediate liner but flow down at the opposite site of the coating knife. For viscosity analysis the predominant shear rate within the coating knife is of essential importance. The shear rate is dependent on the coating velocity. Preliminary experiments showed that coating velocities of 125–159 mm/min lead to optimal films. Therefore, the shear rate was calculated with a velocity of 159 mm/min. For most casted films a WFT of 400–500 μm was chosen due to the drying time and the drug load of the film. This leads to a calculated shear rate of 6 s−1. Viscosity measurements were all performed at this shear rate to imitate the conditions inside the coating knife. Table 2 displays the obtained viscosities of polymeric solutions for the determination of viscosity ranges. All HPMC solutions were tried to be casted at a WFT of 500 μm with a velocity of 159 mm/min. Solutions of 15% HPMC or higher were able to be casted onto the intermediate liner. Lower polymer content lead to a leaking on the front side of the coating knife, so that the polymer solution is not transported onto the intermediate liner. A 15% HPMC solution reveals a dynamic viscosity of 1 Pa*s. Therefore, under usual manufacturing conditions, polymer solutions should have at least a dynamic viscosity of 1 Pa*s. In addition it was observed that higher viscosities lead to reduced spreading of the coating solution onto the intermediate liner, which effects the API content of the film because the API is distributed inside the polymer solution and dosed via the area of the film. A high spreading of the polymer solution leads to a lower wet film thickness and a lower API content per cm2. These effects are not that pronounced during the non-continuous production because the casted film area is larger so that edge effects are not that preponderated. In contrast to the non-continuous film production, the continuous production always requires temperature zones to dry the film in the short time period before the film is coiled up. Especially when working with organic solvents, the drying temperatures might affect the mechanical properties of the films and the content of residual solvents. Therefore, drug-free HPC films were investigated after production dried at different drying conditions. With dynamic viscosities of 3.11 ± 0.01 Pa*s (aqueous solution), 4.50 ± 0.84 Pa*s (acetone/ water) and 3.69 ± 0.03 Pa*s (ethanol/water) all casting film solutions can be casted onto the intermediate liner. All production conditions from 60/80 °C to 100/120 °C led to flexible, transparent films without recrystallization or precipitation of any component. Due to the solvent evaporation, higher amounts of air bubbles can be observed inside films dried at high temperatures, especially at the centre of the film, due to the heat distribution of the heating elements. Therefore, only the weight and not the film thickness was chosen as representative parameter because the measurements with the micrometre screw led to high variations due to the air bubbles. Table 3 displays the results for the weight measurements. Films casted of an acetone/ water mixture weigh less than films casted out of a watery solution. Ethanolic films showed the highest weights but also revealed the highest relative standard deviation (RSD). No uniform drying of ethanolic films was achieved. Fig. 4 shows the mechanical characterization of the films based on casting solutions with different solvent mixtures. Different drying
HPC films
Weight [mg/6 cm2]
Relative standard deviation [%]
Acetone 60/80 °C Acetone 80/100 °C Acetone 100/120 °C Ethanol 60/80 °C Ethanol 80/ 100 °C Ethanol 100/120 °C Water 60/80 °C Water 80/100 °C Water 100/120 °C
20.9 18.5 17.2 29.0 27.7 27.1 23.1 23.4 22.6
± 2.1 ± 0.7 ± 2.1 ± 5.2 ± 7.1 ± 7.4 ± 3.6 ± 3.8 ± 2.3
Fig. 4. Puncture Strength of drug free films casted out of organic solvent mixtures, dried at various drying temperatures (60/80 °C, 80/100 °C, 100/120 °C), n = 6, mean ± SD.
conditions have no significant impact on HPC films cast out of an ethanol/ water mixture, whereas higher drying temperatures lead to a decrease of puncture strength for films casted out of acetone/ water mixtures. This can be explained with the lower boiling point of acetone, which leads to more solvent evaporation and less flexibility and strength at higher drying temperatures. Films based on water show a trend to higher puncture strengths with increasing drying temperatures. For the highest selected drying temperatures, ethanolic films revealed the highest elongation to break (15 ± 4%) followed by aqueous films with an elongation of 10 ± 2%. Films casted out of an acetone/ water mixture showed the lowest elongation to break with 8 ± 3%. The results of the dynamic vapour sorption (Fig. 5) showed no differences on the absorption capacity of different drug free films. Once dried, water sorption remains constant independently of the solvent used for film preparation. Only in the beginning of the measurement, when the relative humidity is at 0%, films based on an acetone/water
Fig. 5. Dynamic vapour sorption of drug-free ODFs casted from film solutions containing acetone, ethanol and water.
289
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
temperatures. A threshold value of 1 Pa*s could be defined for the dynamic viscosity of the casting solutions under common production conditions. Less viscous solutions were not able to be cast onto the intermediate liner but flew down at the opposite site of the coating knife. Drug-free films dried at different drying temperatures show differences in the content of organic solvents and small differences in their mechanical properties. The water absorption capacity is not affected by the selected drying temperatures. Drug-free films casted out of an acetonic solution reveal more air bubbles than films of ethanolic or aqueous casting solutions. The higher the selected drying temperature, the higher the air bubble formation during the drying process. This could lead to variations in content uniformity for API containing films. 3.2. Scale up for an HCT ODF from non-continuous to continuous production
Fig. 6. Loss of drying of drug-free films. Relative humidity was set to 0%. The diagram shows the loss of mass in % over the time.
Previously described results were considered for the scale-up of an orodispersible film containing HCT. The HCT castings solutions were first casted in a non-continuous process simulating the continuous production as far as possible by setting the drying temperatures to 50 °C or 70 °C. The same polymer solutions were subsequently casted utilizing the continuous film coater. A comparison of the resulting films show the differences and the critical process parameters during the production process. ODFs containing HPC as film former led in all cases to transparent and flexible films, irrespectively of the selected drying temperatures. In contrast ODFs made from HPMC resulted in transparent films directly after production, but precipitation was observed after approximately 12 h for the continuous as well as the non- continuous production. Polarised light microscopy reveal no crystals within the film. This leads to the assumption that HPMC precipitates in an amorphous state forming irregular “clouds”. Elevated drying temperatures decreased the amount and sizes of the formed precipitates. The higher the drying temperatures, the lower was the degree of precipitation. In the continuous production mode less precipitation was observed in comparison to the films casted with the Erichsen coater. At drying conditions of 100/120 °C no precipitation was observed at all, but these drying conditions are applicable for the continuous process only. Fig. 7 shows the results of the mechanical properties of HCT films consisting of different polymers, dried at different drying temperatures in either continuous or non-continuous production mode. A distinct difference between HPMC and HPC films can be observed, especially for the non-continuous production. HPC films reveal a lower puncture strength then HPMC films. In contrast to these results, they have a
mixture show lower mass loss compared to films casted out of watery or ethanol/water solutions (Fig. 6). Results of the moisture analysis reveal that with higher drying temperatures less moisture (defined as residual ethanol or acetone in addition to residual water) remains inside the ODFs directly after production (for water based films: 5.4% at 60/80 °C and 3.3% at 100/ 120 °C). Furthermore, at lowest drying temperatures, moisture contents of water based films are the highest (5.4%) whereas the water content decreases when organic solvents were incorporated (4.5% ethanol/ water mixtures and 4.0% acetone/water mixtures). These effects can be explained with the lower boiling point of ethanol and acetone which enables them to evaporate faster than water. Nevertheless, all casted films revealed adequate mechanical properties as they could be coiled up immediately after production. Further handling was feasible without breaking the films. Table 4 displays the results for the gas chromatographic analysis of residual ethanol or acetone of the films directly after production. For films based on an ethanol/ water mixture, the amount of ethanol decreases with increasing drying temperatures. A single film piece contained 2773 ppm ethanol when dried at 60/80 °C, whereas drying at 100/120 °C led to a lower ethanol content of 191 ppm ethanol per single dose. For the films based on an acetone/water solution, no acetone peak could be observed in the chromatogram directly after production. Therefore the content had to be below the detection limit of the method (vial concentration of 0.3 ppm). The low boiling point of acetone led to almost complete evaporation of the solvent during all production conditions. The characterization of the continuous film coating machine showed that coating process is reproducible because the heating and the velocity of the release liner can be precisely adjusted. But the drying of the films is inconsistent during the drying process due to the irregular heat distribution of the heating elements. This leads to centred air bubble formation when working with organic solvents at higher drying
Table 4 Results from GC analysis of residual solvents, mean, n = 2. Solvents basis
Drying condition
Content of residual solvents/ single dose [ppm]
ethanol/water (50/ 50)
60/80 °C 80/100 °C 100/120 °C
2773 910 191
acetone/water (50/ 50)
60/80 °C 80/100 °C 100/120 °C
n.d.a n.d.a n.d.a
a
Fig. 7. Puncture strengths of HCT films made from HPC or HPMC solutions at different temperatures in bench scale and continuous manufacturing mode, n = 6; mean ± SD.
n.d. = acetone content below detection limit.
290
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
Fig. 8. Residual acetone of HCT films based on HPC, dashed line represents the limit for residual solvents (Class 3) according to the Ph. Eur. n = 2, mean. Fig. 9. Content HCT per single dose measured by HPLC, n = 10, mean ± SD.
higher puncture strength with increasing temperatures continuously produced. Therefore, HPC films reveal different mechanical properties in dependency of their production conditions. This effect is more pronounced for the HPC than HPMC films, which exhibit no major changes in puncture strength when varying the drying condition or the mode of the production process. The results of the gas chromatographic analysis to determine the content of residual solvents of HPC films are displayed in Fig. 8. With increasing drying temperatures, the amount of residual acetone decreases. Therefore ODFs produced with the continuous coating machine contain less acetone than films produced in a noncontinuous production mode. This is most probably due to the higher drying temperatures. In comparison to the results obtained for the drugfree HPC films, HCT films reveal a higher acetone content. The presence of an active pharmaceutical ingredient clearly affects the content of residual solvents. According to the Ph.Eur. for adults an amount of less than 5000 ppm per day is acceptable without justification (Ph.Eur., 2016). However, for paediatric purposes lower concentrations may be appropriate. All films prepared using the continuous machine reveal less than 2100 ppm acetone per single dose directly after production and therefore fulfil the requirements regarding the content of residual solvents, whereas the non-continuous produced films show a higher content of residual solvents despite they were analysed 24 h after production to ensure complete dryness of the films. These results illustrate that when working with organic solvents, the scale-up from a noncontinuous to a continuous process is not feasible. Resulting concentrations of residual solvents are not predictable. No conclusions can be drawn from a bench-scale to a continuous process; even if the process settings are known. Different APIs may affect the content of residual solvents. Therefore, for each formulation and every change in process parameters the content has to be determined. The content uniformity of HCT (Fig. 9) in HPMC films is not affected by different drying temperatures neither in the continuous nor in the non-continuous production mode. HPC films generally reveal a lower content of HCT than HPMC films per area of the predefined film piece, especially with higher drying temperatures in the non-continuous production. However, the content uniformity is not affected by this phenomenon (relative SD of 3–6%). ODFs consisting of HPC produced within the continuous production, contain less HCT per single dose than HPMC films (7 mg with HPC as polymer or 13 mg with HPMC as polymer), but are not affected by the different drying temperatures. This confirms that the viscosity of the film solution, in combination with the chosen coating velocity plays an important role for the continuous film casting. A high molecular HPC was chosen as film forming polymer to reduce the spreading effects onto the release liner during the coating process. This leads to a high casting viscosity of 20.4 Pa*s in contrast to the HPMC solutions utilizing a low molecular HPMC, which reveal a viscosity of 2.1 ± 0.08 Pa*s near to the determined threshold value for film casting.
All continuously produced HCT films disintegrate within less than 180 s (max. 14 ± 1 s) and therefore fulfil the requirements of orodispersible dosage forms according to Ph.Eur. (2017). Only the HCT films consisting of HPC dried at 20 °C needed with 67 ± 11 s more than 30 s for disintegration. The comparison of the HCT films in a continuous and non-continuous production mode could show high differences for the content of residual solvents. Continuously produced ODFs reveal a distinct lower amount of acetone than the bench- scale products. This can be explained with the higher drying temperatures. Lab-scale films produced at 70 °C contain a similar amount of residual acetone than the continuously produced ODFs. The production mode also affects the API content of HPC films. Where the HPMC films contain almost the same amount for every production setting, the HCT content is decreasing for HPC films with higher drying temperatures. This can be explained with the spreading of the film solution on the intermediate liner. The higher the selected drying temperature, the less time for spreading and the higher the layer thickness during drying. The mechanical properties are not highly affected by the selection of the production mode. Only the HPC films show differences in puncture strength during bench- scale production. In conclusion, the most critical process parameters for an up-scaling are the viscosity of the coating solution and the selected drying temperatures. 4. Conclusion This study shows that a transfer from a lab-scale to a continuous pilot-scale production is not easy to be performed. Depending on the film-forming polymer more or less differences of film properties due to the casting process can be observed. HPMC films are less affected by the production mode, but display temperature dependent precipitations. HPC films show higher differences depending on the production mode and conditions. Lab-scale production leads to higher drug content but also to higher content of residual solvents. The mechanical strength (expressed as puncture strength) is improved by the continuous production mode. Different requirements on the film solutions also underline the outcome that a transfer from bench-scale to continuous production cannot be easily performed. In conclusion, when transferring film manufacturing from non-continuous to continuous mode, especially the viscosity of the casting solution and the drying temperatures are critical process parameters which cannot be simulated during a lab-scale production on a bench. These parameters have to be selected individually for each product when transferring the process to a pilot-scale or production-scale continuous process. In our opinion, it is even questionable weather discontinuous lab-scale experiments should be conducted if the target should be a production-scale or whether 291
International Journal of Pharmaceutics 535 (2018) 285–292
Y. Thabet, J. Breitkreutz
dissolving films for oral dosage from natural polysaccharides. Materials 3, 4291–4299. Optimags, 2017a. Dr. Zimmermann GmbH, TDS ODV Development Equipment [Accessed 06 September 2017]. http://www.optimags.de/development_equipment.php. Optimags, 2017b. Optimags Coating Machines [Accessed 07 November 2017]. http:// www.optimags.de/advantages_of_optimags.php. Ph.Eur, 2014. European Pharmacopoeia Comission, Oromucosal Preparations, European Pharmacopoeia. European Directorate for the Quality of Medicines & Healthcare (EDQM), Strasbourg, France. Ph.Eur, 2016. European Pharmacopoeia Commission, Residual Solvents, European Pharmacopoeia. European Directorate for the Quality of Medicines & Healthcare (EDQM), Strasbourg, France. Ph.Eur, 2017. European Pharmacopoeia Comission, Orodispersible Tablets, European Pharmacopoeia. European Directorate for the Quality of Medicines & Healthcare (EDQM), Strasbourg, France. Pires, M.A.S., Souza dos Santos, R.A., Sinisterra, R.D., 2011. Pharmaceutical composition of hydrochlorothiazide: β-cyclo-dextrin: preparation by three different methods, physico-chemical characterization and in vivo diuretic activity evaluation. Molecules 16, 4482–4499. Preis, M., 2015. Orally disintegrating films and mini-tablets—innovative dosage forms of choice for pediatric use. AAPS PharmSciTech 16, 234–241. Preis, M., Gronkowsky, D., Grytzan, D., Breitkreutz, J., 2014a. Comparative study on novel test systems to determine disintegration time of orodispersible films. J. Pharm. Pharmacol. 66, 1102–1111. Preis, M., Knop, K., Breitkreutz, J., 2014b. Mechanical strength test for orodispersible and buccal films. Int. J. Pharm. 461, 22–29. Preis, M., Woertz, C., Schneider, K., Kukawka, J., Broscheit, J., Roewer, N., Breitkreutz, J., 2014c. Design and evaluation of bilayered buccal film preparations for local administration of lidocaine hydrochloride. Eur. J. Pharm. Biopharm. 86, 552–561. Visser, J.C., Woerdenbag, H.J., Hanff, L.M., Frijlink, H.W., 2017. Personalized medicine in pediatrics: the clinical potential of orodispersible films. AAPS PharmSciTech 18, 267–272. Woertz, C., Kleinebudde, P., 2015. Development of orodispersible polymer films containing poorly water soluble active pharmaceutical ingredients with focus on different drug loadings and storage stability. Int. J. Pharm. 493, 134–145.
continuous pilot-scale equipment should be used from the beginning. Funding sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. References Bansal, S., Bansal, M., Garg, G., 2013. Formulation and evaluation of fast dissolving film of an antihypertensive drug. Int. J. Pharm. Chem. Biol. Sci. 3, 1097–1108. Buanz, A.B., Belaunde, C.C., Soutari, N., Tuleu, C., Gul, M.O., Gaisford, S., 2015. Ink-jet printing versus solvent casting to prepare oral films: effect on mechanical properties and physical stability. Int. J. Pharm. 494, 611–618. Cilurzo, F., Cupone, I., Minghetti, P., Selmin, F., Montanari, L., 2008. Fast dissolving films made of maltodextrins. Eur. J. Pharm. Biopharm. 70, 895–900. Dixit, R., Puthli, S., 2009. Oral strip technology: overview and future potential. J. Control. Release 139, 94–107. El-Setouhy, D.A., El-Malak, N.S.A., 2010. Formulation of a novel tianeptine sodium orodispersible film. AAPS PharmSciTech 11, 1018–1025. Hoffmann, E.M., Breitenbach, A., Breitkreutz, J., 2011. Advances in orodispersible films for drug delivery. Expert Opin. Drug Deliv. 8, 299–316. ICH, 2009. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH), Pharmaceutical Development, Q8 (R2). ICH, 2015. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH), Validation of Analytical Procedures: Text and Methodology Q2 (R1). Jani, R., Patel, D., 2015. Hot melt extrusion: an industrially feasible approach for casting orodispersible film. Asian J. Pharm. Sci. 10, 292–305. Kalepu, S., Nekkanti, V., 2015. Insoluble drug delivery strategies: review of recent advances and business prospects. Acta Pharm. Sin. B 5, 442–453. Murata, Y., Isobe, T., Kofuji, K., Nishida, N., Kamaguchi, R., 2010. Preparation of fast
292