Determinants of zero-order release kinetics from acetaminophen-layered Suglet® pellets, Wurster-coated with plasticized Aquacoat® ECD (ethyl cellulose dispersion)

Journal Pre-proofs Determinants of zero-order release kinetics from acetaminophen-layered Suglet® pellets, Wurster-coated with plasticized Aquacoat® ECD (ethyl cellulose dispersion) Sandeep Kaur, Sowmya Sivasankaran, Erika Wambolt, Sriramakamal Jonnalagadda PII: DOI: Reference:

S0378-5173(19)30918-4 https://doi.org/10.1016/j.ijpharm.2019.118873 IJP 118873

To appear in:

International Journal of Pharmaceutics

Received Date: Revised Date: Accepted Date:

16 August 2019 5 November 2019 12 November 2019

Please cite this article as: S. Kaur, S. Sivasankaran, E. Wambolt, S. Jonnalagadda, Determinants of zero-order release kinetics from acetaminophen-layered Suglet® pellets, Wurster-coated with plasticized Aquacoat® ECD (ethyl cellulose dispersion), International Journal of Pharmaceutics (2019), doi: https://doi.org/10.1016/ j.ijpharm.2019.118873

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2019 Elsevier B.V. All rights reserved.

Determinants of zero-order release kinetics from acetaminophen-layered Suglet® pellets, Wurster-coated with plasticized Aquacoat® ECD (ethyl cellulose dispersion).

Sandeep Kaur1,2, Sowmya Sivasankaran1, Erika Wambolt2, Sriramakamal Jonnalagadda1* 1Department

of Pharmaceutical Sciences, Philadelphia College of Pharmacy, University of the Sciences, 600 S 43rd Street, Philadelphia, PA 19104, USA.

2Global

Technical Operations, Johnson & Johnson, Fort Washington, PA

*Corresponding author: Sriramakamal Jonnalagadda, Ph.D Professor, Department of Pharmaceutical Sciences, Philadelphia College of Pharmacy, University of the Sciences, 600 S 43rd Street, Philadelphia, PA 19104, USA. Email: [email protected] Tel.: +1 (215) 596-8942, Fax: +1 (215) 895-1161 1

ABSTRACT The potential for zero-order drug-release was evaluated for ethyl cellulose (EC) coated, acetaminophen-layered sugar pellets (Suglets®) of mesh size 18/20 (850–1000 μm). To determine optimal plasticizer/pore-former concentrations for EC films, solvent-cast Aquacoat® ECD (ethyl cellulose dispersion) films were prepared with 0-50% w/w ratios of two triethyl citrate (TEC) or triacetin (TA). Characterization studies showed that films with excipient concentrations ≥ 20% were homogenous, mechanically strong at room temperature (Young’s modulus of 25-35 MPa), and have a glass transition (Tg) in the range of 20-45C. Based on these results, a working range of 20-50% weight concentrations was selected for drug release studies. Suglets® were layered with acetaminophen (APAP) using Wurster Glatt GPCG-3 to yield roughly 10% w/w coating (controls). The Controls were coated using the same Wurster process with Aquacoat® ECD containing 20-50% w/w of TEC or TA. Samples were removed periodically at 311% weight gain, to evaluate impact of weight gain, and consequently film-formation, on drug release. Dissolution was monitored over a period of 12 hours in a media consisting of simulated gastric fluid (first two hours), followed by simulated intestinal fluid. The controls showed near 100% release within the first 30 minutes, indicating the value of EC-coating to achieve controlled release. Dissolution release profiles showed that TEC is more effective than TA as a plasticizer and pore-former, as linear profiles were apparent at lower concentrations and % weight gain. For a quantitative evaluation of these results, linear regression was fitted to all cumulative release profiles, and R-square values examined as a function of excipient concentration and % weight gain. The corresponding response surface plots and the second order regression were shown to aid in optimization. The design space for zero-order release was represented as contour plots between excipient concentration and % weight gain.

Keywords: Zero-order Release, Ethyl Cellulose, Thermal analysis, Wurster Coating, Triethyl Citrate (TEC), Triacetin, Response surface, Contour plots, dissolution testing.

2

BACKGROUND Multiparticulate drug/polymer layered pellets are frequently used in designing oral sustainedrelease dosage forms. Such dosage forms combine a single dose of the selected therapeutic agent or drug into discrete pellets measuring approximately 0.05-2.00 mm in diameter.1, 2 These pellets are typically manufactured by spheronizing a powder blend of the drug and excipients, or by layering inert cores with a drug/excipient solution or suspension.1 Pellets are known to reduce variability in transit times (food effect), and can therefore optimize drug absorption and bioavailability.3-5 As a dosage form, pellets comparatively exhibit less dose-dumping risk compared to coated tablets or capsules, because a single dosage of pelletized drug is comprised of thousands of smaller sub-units causing failure of all pellets to be statistically improbable.2 Finally, the availability of inert sugar spheres of uniform shape, surface area, high density, low friability, and low particle-size variation, combined with scalable drug/polymer-coating technologies, makes coated-pellets a feasible technology that offers great flexibility in formulating oral-dosage forms.3, 6 Ethyl cellulose (EC) is a frequently used hydrophobic polymer in controlled-release pellet coatings.7-12 This polymer can be dissolved in organic solvents or emulsified in aqueous-based dispersions.13 The use of aqueous dispersions is ideal over organic solvent-based solutions due to reduced toxicity and flammability concerns.13, 14 Aquacoat® ECD (26% w/w EC, 2.4% w/w cetyl alcohol, and 1.3% w/w sodium lauryl sulfate (SLS)), supplied by FMC Biopolymer (Philadelphia, PA), is a commercially available ethyl cellulose aqueous dispersion, used extensively for the development of pharmaceutical dosage forms. Polymeric films formed from Aquacoat® ECD tend to be rigid and inflexible at room temperature. This is due to a relatively higher glass transition temperature (Tg) of about 89°C.15, 14 Plasticizers are therefore typically added to lower the Tg and impart flexibility to Aquacoat®-based polymeric films.14, 16 Apart from plasticizers, pore-forming agents are also needed to ensure drug release from EC-coated formulations. Examples of EC-compatible poreforming agents include hydrophilic polymers such as PVA-PEG, HPMC, and carrageenan.17-22 The use of polymeric excipients as pore-forming agents has been shown to cause destabilization of aqueous ethyl cellulose dispersions resulting in unpredictable drug-release profiles.23-25 The destabilization is a consequence of macro-phase separation between immiscible domains of the hydrophobic EC films and the hydrophilic pore formers. This paper evaluates the use of two small-molecule hydrophilic excipients, triethyl citrate (TEC) and triacetin, to serve the dual role of plasticizer and pore-former for EC films prepared using Aquacoat® dispersions. The elimination of hydrophilic polymer-based pore-formers is expected to improve stability and homogeneity during preparation and storage, while the use of a single plasticizing/poreforming excipient would contribute to lower manufacturing cost. Acetaminophen (APAP) was used as a model drug in these studies. APAP is a commonly used drug, with a wide range of applicability in oral dosage forms. A relatively short half-life of about 4 hours has prompted the development of numerous sustained release oral dosage forms for APAP.26 Zero-order release of APAP from hot-melt coated, wax-layered pellets has previously been reported in literature.27 In this research, polymeric films were cast from Aquacoat® ECD 3

using TEC or triacetin at concentrations ranging from 0 to 50% w/w. The thermomechanical properties of these dried polymeric films were evaluated using thermal analysis, mechanical testing, and scanning electron microscopy. Subsequently, sugar spheres (Suglets®) were layered with APAP, then coated with Aquacoat® ECD containing 20-50% w/w of TEC or triacetin using a Wurster coater. The pellets were periodically removed at coating weight-gains ranging between 3-11% w/w. The release of drug from Aquacoat® ECD films was evaluated to determine the effect of plasticizer/pore-former type, concentration, and coating weight gain.

MATERIALS AND METHODS Materials: Aquacoat® ECD-30 with a total solid concentration of 29–32% w/w is an aqueous ethyl cellulose-based dispersion from FMC (FMC Corporation, Philadelphia, USA). TEC and triacetin were obtained from Thermo Fisher Scientific Chemicals Inc. (Sparks, NV, USA). Sugar pellets— uglets®: 18/20 mesh (size: 850–1000 μm) (Colorcon Inc, West Point, PA, USA); hydroxypropyl methylcellulose, HPMC (Methocel® E5, Colorcon Inc, West Point, PA, USA), and acetaminophen (APAP) (Mallinckrodt Inc., St. Louis, MO, USA) were used for these studies. Methods: Preparation of polymeric films: Preparation of Polymeric Films: TEC or triacetin (weight ratios of 0, 5, 10, 20, 30, 40, and 50% w/w based on total solid content) were added to 30 mL of Aquacoat® ECD-30 dispersion and mixed using a magnetic stir bar for 45 to 60 minutes. Subsequently, purified water was added to attain a 15% w/w solids suspension. About 35 mL of this mixture was transferred into 5-inch diameter aluminum pans, followed by drying in an oven at a temperature setting of 40°C for 24 hours to obtain thin films. The resulting polymeric films (thickness 0.85  0.05 mm) were manually removed from the pans and dried for an additional 24-hour period at ambient conditions. The films were cut into rectangular strips of 31.3 cm for further characterization and testing. Thermal Characterization: Thermal characteristics of polymeric films were performed using a Discovery DSC2500 (TA Instruments) equipped with a Refrigerated Cooling System (RCS90). The prepared polymeric film samples weighing approximately 5-7 mg were each placed inside an aluminum crucible and sealed using a crimping press. Samples were heated from room temperature to 150°C at 10°C /min. After cooling to 0°C, they were re-heated to 150°C at the same rate. Thermograms corresponding to the second heating cycle (H2) were evaluated to examine the effects of polymer: excipient compatibility. Results were analyzed using the Trios software. Mechanical Characterization: The mechanical properties of the prepared polymeric films were evaluated using the Instron 3360 Universal Testing system (Norwood, MA) mounted with a 5 N load cell. The rectangular cut film specimens (3×1.3 cm) were held between two grips. The extension speed for the moving grip was maintained at 10 mm/min. The resulting stress/strain 4

plots were evaluated for tensile strength, percent elongation, and Young’s modulus using Instron Bluehill testing software based on the equations below: 28-30 𝐓𝐞𝐧𝐬𝐢𝐥𝐞 𝐬𝐭𝐫𝐞𝐧𝐠𝐭𝐡 =

𝐋𝐨𝐚𝐝 𝐚𝐭 𝐟𝐚𝐢𝐥𝐮𝐫𝐞 𝐟𝐢𝐥𝐦 𝐭𝐡𝐢𝐜𝐤𝐧𝐞𝐬𝐬 × 𝐟𝐢𝐥𝐦 𝐰𝐢𝐝𝐭𝐡

𝐏𝐞𝐫𝐜𝐞𝐧𝐭 𝐞𝐥𝐨𝐧𝐠𝐚𝐭𝐢𝐨𝐧 = 𝐘𝐨𝐮𝐧𝐠′𝐬 𝐦𝐨𝐝𝐮𝐥𝐮𝐬 =

𝐈𝐧𝐜𝐫𝐞𝐚𝐬𝐞 𝐢𝐧 𝐟𝐢𝐥𝐦 𝐥𝐞𝐧𝐠𝐭𝐡 × 𝟏𝟎𝟎 𝐈𝐧𝐢𝐭𝐢𝐚𝐥 𝐟𝐢𝐥𝐦 𝐥𝐞𝐧𝐠𝐭𝐡

𝐒𝐥𝐨𝐩𝐞 𝐟𝐢𝐥𝐦 𝐭𝐡𝐢𝐜𝐤𝐧𝐞𝐬𝐬 × 𝐜𝐫𝐨𝐬𝐬 ― 𝐡𝐞𝐚𝐝 𝐬𝐩𝐞𝐞𝐝

(𝟏) (𝟐) (𝟑)

Environmental Scanning Electron Microscopy: The surface characterization of the polymeric films was performed using an FEI Quanta 250 environmental scanning electron microscope (ESEM) (Thermo Fisher Scientific, Hillsboro, Oregon, USA). Film samples were placed onto the conductive aluminum specimen stub with adhesive carbon discs. The stage with the film sample was raised at 10 mm working distance of the polar piece of the microscope, and images were taken from the surface with the voltage set at 10 kV.

Preparation of Coated pellets: Drug (APAP) layering: Acetaminophen was layered onto sugar pellets using an 18% (w/v) suspension in water containing 5% w/w HPMC (Methocel® E5) with a Wurster processor (Glatt GPCG-3) to achieve a 10% w/w APAP content. The drug-layered pellets were manufactured using the formulation shown in Table 1. Table 1: Drug Layer Formulation Ingredients APAP Suglets Methocel E5 Total

%w/w 10 85 5 100

Grams per Batch 350 2975 175 3500

The layering process parameters were: batch size 3500 g, inlet air temperature 70°C, product temperature 45°C, air flow 45 cfm, nozzle diameter 1.0 mm, spray rate 15 g/min, and spray pressure 1.2 bar. Following the drug layering process, the layered pellets were dried for 10 min. For the drug layering process, the final binder (HPMC) concentration and processing parameters were selected based on previously conducted experimental runs exploring different binder concentrations and process parameter targets.

5

Aquacoat® ECD coating: The drug-layered pellets were further coated with Aquacoat® ECD using excipients (TEC or triacetin) at concentrations of 20, 30, 40 or 50% w/w (based on ethyl cellulose content). Excipients were incorporated into Aquacoat® ECD by stirring for 30-45 minutes, and purified water was added to obtain a 15% w/w solids/suspension. Roughly 10-15 g samples were taken at 3, 5, 7, 9, and 11% weight gain to study the effect on drug-release. Coating operation was executed in a Wurster Glatt GPCG-3 model to achieve certain weight gain using the following process parameters: batch size 1000 g; inlet temperature 50 °C, product temperature 36 °C, air flow 35 cfm, nozzle diameter 1.2 mm, and spray rate 8 g/min. Subsequently, the polymer-coated pellets were dried for 10 min at 40ºC. Similar to the drug layering process, the final processing parameters for Aquacoat® ECD coating were selected based on previously conducted experimental runs exploring different process parameter targets and ranges. Surface Characterization-Environmental Scanning Electronic Microscopy (ESEM): The surface characteristics of the pellets were examined using a FEI Quanta 250 environmental scanning electron microscope (Thermo Fisher Scientific, Hillsboro, Oregon, USA). The microscope creates detailed high-resolution imaging in a digital format. Pellet samples were positioned on carbon adhesive tabs placed on the top of the aluminum specimen stubs. The pin on the underside of the specimen stub mount was inserted into the opening in the top of the microscope stage. The stage with pellet samples was raised to 10 mm height, and high-resolution images were taken under low vacuum mode using a scan time of 100µs. Magnification, brightness, sharpness, and focus were adjusted to obtain a clear image. Drug Release Study: Drug release from the pellets was studied using the dissolution apparatus Model 6300 (Distek, North Brunswick, NJ) using the USP <711> paddle (apparatus II) method, at a stirring speed of 75 rpm. The quantity of pellets used for the study was 6.5 grams. The dissolution medium at 30, 60, 90, and 120 min was 750 mL of 0.1 N HCl (simulated gastric fluid). After the 120 min testing interval, 250 mL of 0.20 M tribasic sodium phosphate (simulated intestinal fluid) was added to each dissolution vessel, and the pH of the medium was adjusted to 6.8 using 1N HCL or NaOH. The drug content in each dissolution sample was analyzed using an in-line UV Fiber Optic Dissolution System (OPT- DISSTM) (Leap technologies Inc, Morrisville, NC) at the wavelength of 243 nm corresponding to λmax for APAP. Statistical analysis: Mechanical strength testing of the films was done in triplicate (n = 3). The tensile strength, % elongation, and elastic moduli were analyzed for significant difference between the two plasticizers, TEC and Triacetin, using a two-way ANOVA with replication. Statistical significance was established at p < 0.05. Drug release studies were conducted in triplicate. Data from the drug release study was graphed and analyzed using Microsoft® Excel to generate linear regression fits. The corresponding zero-order release constant and R2 values were subsequently re-analyzed using Minitab® 18.1 to generate contour plots, which were used to depict the effect of plasticizer concentration and coating weight gain on membranecontrolled release and release rate. Response surface optimization was also done using Minitab to obtain optimal plasticizer and coating weight gain conditions for a slow, membranecontrolled release. 6

RESULTS Thermal Characterization of Polymeric Films: Figures 1a and 1b show DSC thermograms for Aquacoat® ECD-30 films containing TEC and Triacetin at concentrations up to 50% w/w. The thermograms showed two distinct endothermic peaks in the temperature range of 20-40⁰C and 75-85⁰C. The DSC thermograms with excipient concentrations > 20% w/w showed a broad endothermic transition in the range of 20-40⁰C that progressively increased in area with excipient concentration. The onset of this endotherm was in the temperature range of 25-30⁰C for films containing excipient concentration at 20% w/w, with higher concentrations causing the onset to occur at lower temperatures. This transition was also associated with a significant lowering in the baseline following the transition, a clear indication of Tg, and possible plasticization of Aquacoat® films for excipient concentrations > 20% w/w. All DSC thermograms for both excipients also showed a consistent endothermic transition in the range of 78-85⁰C. The enthalpy of this transition ranged from 0.9 to 0.3 J/g (Table 2). As evident from Table 2, there was a steady decline of this enthalpy with decreasing weight ratio of Aquacoat®. This transition was, therefore, identified to be due to the original components of Aquacoat® ECD.

Figure 1(a): DSC Thermograms Showing the Effect of TEC Concentration on Aquacoat® ECD-30 Films

7

Figure 1(b): DSC Thermograms Showing the Effect of triacetin Concentration on Aquacoat® ECD-30 Films

Table 2: Enthalpy values for endothermic peaks for Aquacoat®: excipient Films Excipient Conc. (% w/w) 0 5 10 20 30 40 50

H of 20-40ºC endotherm (J/g) TEC Triacetin ND ND ND ND ND ND 1.06 0.85 1.63 2.12 2.89 7.63 6.65 8.70

H of 75-85ºC endotherm (J/g) TEC Triacetin 0.73 0.73 0.89 0.89 0.83 0.76 0.67 0.60 0.52 0.55 0.41 0.40 0.33 0.38

ND: Not Detected

Mechanical Characterization of Polymeric Films: Table 3 shows the comparative mechanical testing results of the excipient polymeric films with TEC and triacetin at concentrations corresponding to the DSC thermograms. An accurate determination was only possible for TEC and triacetin-containing films of concentration ≥ 20%. Films with excipient concentration < 10% lacked the consistency to enable mounting and mechanical testing. All films prepared from both TEC and triacetin in the range of 30-50% excipient weight showed comparable mechanical properties. Compared to films prepared with 20% excipient concentration, these films showed a lower tensile strength and Youngs’ modulus, and correspondingly higher elongation. 8

Table 3: Comparative Mechanical Properties of the Excipient Polymeric Films with TEC and Triacetin at Different Concentrations Tensile Elongation Young’s Strength (%) * Modulus Mpa MPa * 1 0 ND ND ND 1 5 ND ND ND 1 10 ND ND ND 20 1.43 (±0.72) 9.11 (±0.78) 29.12 (±0.99) Triethyl 30 1.18 (±0.47) 11.56 (±0.50) 25.67 (±0.45) Citrate 40 1.10 (±0.39) 11.20 (±0.18) 24.20 (±0.72) 50 1.12 (±0.37) 10.19 (±0.14) 24.01 (±0.96) 1 5 ND ND ND 1 Triacetin 10 ND ND ND 20 1.67 (±0.98) 10.19 (±0.87) 34.54(±1.10) 30 1.48 (±0.84) 12.32 (±1.02) 29.16 (±1.00) 40 1.42 (±0.92) 13.56 (±1.18) 28.08 (±1.14) 50 1.41 (+0.66) 13.01 (±0.90) 28.07 (±0.90) 1Values could not be determined accurately due to the highly brittle nature of the films *significantly different between the two excipients, Triethyl Citrate and Triacetin. Excipient

Concentration (% w/w)

Surface Characterization of Polymeric Films: Figures 2 and 3 depict ESEM images of the Aquacoat® excipient films. Aquacoat® films cast with TEC concentrations of ≤ 20% w/w showed numerous visual cracks that were not observed in films at higher TEC concentrations. ESEM analysis of pure Aquacoat® films showed macrophase separation in the form of needle-shaped structures. In contrast to Aquacoat® TEC-containing films, ESEM micrographs of triacetincontaining films were non-homogeneous at all excipient concentrations. Aquacoat films containing 5 and 10% w/w of triacetin contained large, linear cracks. At triacetin concentrations, ≥ 20% w/w, the appearance of macrophase separation was more in the form of granular nodules. Overall, the ESEM results indicate that the preparation of homogeneous films from Aquacoat® requires the addition of plasticizers in concentrations of 20% w/w or higher.

9

Figure 2: Scanning Electron Micrographs of Polymeric Films at 500 m (a) pure Aquacoat® (b) 5% w/w TEC, (c) 10% w/w TEC, (d) 20% w/w TEC, (e) 30% w/w TEC, (f) 50% w/w TEC (a) (b) (c)

(d)

(e)

(f)

Figure 3: Scanning Electron Micrographs of Polymeric Films at 500 m (a) pure Aquacoat®, (b)5%w/w Triacetin, (c)10%w/w Triacetin, (d)20%w/w Triacetin, (e) 30% w/w Triacetin, (f) 50% w/w Triacetin (a) (b) (c)

(d)

(e)

(f)

Drug Release from Uncoated APAP Layered Pellets: For the drug layered pellets, 6.5 grams of pellets with a drug content of approximately 0.65 g (650 mg) based on 10% w/w concentration 10

were used for dissolution testing. Table 4 below shows the drug release from uncoated drug pellets. Results conform to the criteria established for immediate release dosage forms (> 80% release in 30 minutes). Table 4: Drug Release from Drug Layered Pellets Time Point (min) 5

Drug Released (grams) 0.14

Amount of Drug Release (%) 22% (+0.53)

15

0.40

62%(+0.23)

30

0.63

97%(+0.20)

Drug release from Polymer Coated Beads with TEC and Triacetin: Figures 4 and 5 shows the effect of weight gain on the release of drug from EC coated pellets. The drug release for the polymer-coated beads was performed in triplicate, with a calculated standard deviation < 3% of the mean. The drug-release profiles indicate that the use of 20% w/w of TEC and triacetin satisfied the immediate drug release (> 80% released within 30 minutes) at almost all weight gain percentages, suggesting incomplete film formation in these pellets. The only anomaly was observed at 11% weight gain for 20% w/w concentration of TEC, where a linear drug release profile occurred over six hours. For all other concentrations, the duration of drug release appeared to increase progressively with increasing weight gain. The relative drug-release data at different coating weight gains show that the release rate decreases with an increase in the coating weight gain. That is, the thicker the film coat, the longer the diffusion path-length, and slower the release rate. In addition, as coating-film uniformity is achieved at higher coating weight gain, there is no significant effect of excipient concentrations (TEC or triacetin) beyond 30% concentration on drug release due to the possible saturation of the coating polymer as indicated by the previously conducted polymeric free-film studies.

11

Figure 4: Effect of Coating Weight Gain on Drug Release from Drug Layered Pellets Coated with Aquacoat® ECD Dispersion using TEC at a) 20% w/w b) 30% w/w, c) 40% w/w and d) 50% w/w Concentrations.

a) TEC 20%

100

80 % Released

% Released

80 60 40

60 40

20

20

0

0 0

3

6 Time (Hours)

9

0

12

c) TEC 40%

100

3

6 Time (Hours)

9

12

9

12

d) TEC 50%

100 80 % Released

80 % Released

b) TEC 30%

100

60 40

60 40 20

20 0

0 0

3

6 Time (Hours)

9

12

0

3

6 Time (Hours)

12

Figure 5: Effect of Coating Weight Gain on Drug Release from Drug Layered Pellets Coated with Aquacoat® ECD Dispersion using Triacetin at a) 20% w/w b) 30% w/w, c) 40% w/w and d) 50% w/w Concentrations.

b) Triacetin 30%

a) Triacetin 20%

100

100 80 % Released

% Released

80 60 40

40

20

20

0

0 0

3

6 Time (Hours)

9

0

12

3

c) Triacetin 40%

100

d) 100

80

6 Time (Hours)

9

12

Triacetin 50%

80 % Released

% Released

60

60 40 20

60 40 20

0

0 0

3

6 Time (Hours)

9

12

0

3

6 Time (Hours)

9

12

Surface Characterization of the pellets: Figure 6 shows the effect of TEC or triacetin concentration on Aquacoat® ECD-30 coated pellets. APAP-layered pellets coated with Aqueous ECD with TEC or triacetin at 20% w/w concentration had visible cracks. However, APAP layered pellets coated with Aqueous ECD using TEC at a concentration of 30, 40, and 50% w/w showed a continuous film with a more uniform appearance. In comparison to TEC, triacetin showed poor film formation with a rougher surface and cracks at all evaluated concentrations.

13

Figure 6: Scanning Electron Micrographs of the Aqueous Ethyl Cellulose Dispersion (Aquacoat®) Coated Drug Layered Pellets with a) TEC and b) Triacetin at i) 20%, ii) 30%, iii) 40% and iv) 50% w/w concentrations

a.i)

b.i)

a.ii)

b.ii)

a.iii)

b.iii)

a.iv)

b.iv)

14

DISCUSSION: Choice of excipients:

Figure 7 shows the structure of TEC and Triacetin. Both excipients are colorless liquids with comparable physical properties. TEC has a MW of 276.285 g/mole, with a melting point of – 55⁰C and boiling point of 294⁰C.31 Triacetin has a MW of 218.208 g/mole, the melting point of – 78⁰C, and boiling point of 280⁰C.32 The solubility in water for these compounds is also comparable at roughly 65 and 58 g/liter, respectively, at room temperature. The partitioning extent of TEC and triacetin in aqueous ethyl cellulose dispersions have been quantified previously by calculation of association coefficients.33 The association coefficients for TEC and triacetin for ethyl cellulose polymer were calculated to be 5.70 and 3.14, respectively.33 The chemical structures of these excipients are shown below. Figure 7: Chemical Structures of (a) Triethyl Citrate, and (b) Triacetin.

(a)

(b)

Physical properties of Aquacoat® films: The DSC thermograms of Aquacoat® films showed two distinct endothermic peaks at 20-40⁰C and 75-85⁰C. The endotherm in the temperature range of 20-40⁰C was attributed to the excipient (TEC or triacetin), based on the observation of a progressive increase in peak enthalpy at excipient concentrations > 20% w/w. For excipient concentrations below 20% w/w, the complete absence of any endotherm in this temperature range suggests that the excipients completely integrate into (miscible with) the ethyl cellulose matrix at the molecular level. The melting peak in the range of 78-85⁰C was close to the literature reported Tg of Aquacoat® at 89⁰C.14, 15 Based on our DSC data, this peak appears to correspond to melting endotherm rather than a glass transition, and may be attributed to one or more of several components in the Aquacoat® dispersion used for film-formation. The DSC results, especially the endothermic peaks in the lower concentration range, suggest that the “plasticity” effects of the excipient on Aquacoat® films would be maximized at roughly 15

20% w/w, whereas excipient contribution to film-porosity would occur at concentrations > 20% w/w. Finally, Scanning Electron Microscopy clearly showed inhomogeneous films with numerous cracks at lower concentrations. Compared to triacetin, TEC-containing Aquacoat® films appeared more homogeneous, with fewer incidence of macrophase separation. These results are consistent with TEC having a higher association constant with ethyl cellulose compared to triacetin. Drug release characteristics of coated pellets: Typical membrane controlled (or reservoir) systems are expected to release the drug at a zeroorder rate, after an initial lag time. The lag-time is proportional to h2/6D, where h is the thickness of the membrane, and D is the diffusion coefficient of the relevant species. The release data in our studies did not show a lag-time, but rather a burst. Burst release is not atypical in coated-pharmaceutical pellets and may be a consequence of incomplete filmformation, or film-saturation with the active drug moiety. The latter phenomenon, i.e., membrane saturation with the drug, frequently occurs upon storage and may be controlled by minimizing residual moisture during storage. Figure 8 shows surface plots obtained from modelling the release data. The independent variables in these plots were plasticizer concentration ranging from 30% to 50% w/w of Aquacoat®, and coating weight gain ranging from 3% to 11% w/w. Regression coefficients (R2) obtained from the zero-order release equation and zero-order release rate (k) were the response or dependent variables. Figure 8: Surface plot of Fit for zero-order release Vs. Weight gain and concentration of i) TEC (left) and Triacetin (right).

Surface Plot of R-squared vs Weight gain, TEC Concentration

Surface Plot of R-squared vs Weight gain, Triacetin Concentration

1 .0 1.0

R -S quared

0.9

0 .9

R-squared

0 .8

0 .8

0.7

0.7

10 8 30

6 40

TEC Concentration

Weight gain

4 50

4 6

50 0

8

40

Triacetin Concentration

10

Weight gain

30

Equations for Aquacoat films with TEC: The multivariate Response Surface Regression equations corresponding to the 3D plots (figure 8) are shown below:

16

APAP Release rate = 65.6* - 1.94(TEC Conc.) - 4.39(Wt. gain)* + 0.02(TEC Conc.)2 + 0.08(Wt. gain) 2 + 0.05 (TEC Conc.*Weight gain); (R-square: 0.76, P = 0.012) R-square = 0.099* + 0.21 (Wt. gain)* + 0.01*(Wt. gain) 2; (R-square: 0.85, p = 0.001) Equations for Aquacoat® films with triacetin: APAP Release rate= 522 + 13.4*(TA Conc.) – 165.3*(Wt. gain) - 0.33(TA Conc.)2 + 6.04* (Wt. gain) 2 + 1.36(TA Conc.*Wt. gain)*; (R-square: 0.89, p = < 0.001) R-square = 0.99 + 0.02(TA Conc.) - 0.1921(Wt. gain) + 0.01(Wt. gain) 2; (R-square: 0.35, p > 0.48) * Statistically significant terms (p<0.05). Figure 9 shows an illustration of the use of an optimization function, whereby a target excipient (TEC) concentration and percent weight gain was predicted that would maximize the R-square value of a linear fit, while minimizing release-rate. The intersection of the solid-red vertical line and the dashed horizontal blue line shows the predicted target. Figure 9: Optimized plasticizer concentration and coating weight gain for zero-order release of drug from controlled release pellets for TEC. Optimal High D: 0.9841 Cur Predict Low

TEC Conc 50.0 [41.5152] 30.0

Weight g 11.0 [8.9798] 3.0

R-square Targ: 1.0 y = 0.9939 d = 0.98401

Release Minimum y = 2.8034 d = 0.98424

The second part of the release mechanism in membrane-controlled systems is the drug-release rate. Such a rate has relevance when the data fits a linear profile. To better understand the cumulative effects of weight gain and excipient concentration, the results from figures 4 and 5 were fit using linear regression, with the regression coefficients (R-square values of fits) and zero-order rate constants (for R-square > 0.9) shown as contour plots in Figure 10. The lightly shaded regions in the 2D contour plots show regions in weight-gain/excipient-concentration

17

design space, where zero-order drug release may be obtained. The contour plots show that the TEC-coating is predictable, and highly consistent, compared to Triacetin-coating. Figure 10a: Contour Plot showing R-squared values corresponding to a zero-order fit. (i)TEC Concentration (left) and (ii) Triacetin Concentration (right). Contour Plot of Fit vs Weight gain, Triacetin Concentration

Contour Plot of R-squared vs Weight gain, TEC Concentration R-squared < 0.65 0.65 – 0.75 0.75 – 0.85 0.85 – 0.95 > 0.95

10

% Coating Weight Gain

9 8 7 6 5

11

R-squared < 0.65 0.65 – 0.75 0.75 – 0.85 0.85 – 0.95 > 0.95

10 9

% Coating Weight Gain

11

8 7 6 5 4

4 3 30

35

40

45

3 30

50

35

40

45

50

Triacetin Concentration (% w/w)

TEC Concentration (%w/w)

Figure 10(b) shows the zero-order rate constants as a function of % weight-gain and Triacetin concentration. Overall, the graphs indicate that the release rate declines consistently with weight-gain, for both TEC and triacetin. Figure 10b: Contour Plots showing zero-order release rate constants (i)TEC Concentration (left) and (ii) Triacetin Concentration (right).

11

Release rate < 3.0 3.0 – 4.5 4.5 – 6.0 6.0 – 7.5 > 7.5

% Coating Weight Gain

10

9

8

7

Contour Plot of Release rate vs Weight gain, Triacetin Concentration 7

% Coating Weight Gain

Contour Plot of Release rate vs Weight gain, TEC Concentration

Release rate < 100.0 100.0 – 150.0 150.0 – 200.0 200.0 – 250.0 > 250.0

6

5

4

6

5 30

35

40

TEC Concentration (%w/w) Results exclude rows where Fit < 0.90.

45

50

3 30

32

34

36

38

40

Triacetin Concentration (% w/w) Results exclude rows where Fit < 0.90.

The analysis shown in figures 10(a) and 10(b) allows formulators to identify the weight gain required and excipient concentration while ensuring complete film-formation. The results were also modeled using Response Surface analysis to develop an equation with predictive value.

CONCLUSION: This research demonstrates that Aquacoat® ECD when modified with sufficient triethyl citrate or triacetin, can be solvent-casted into plasticized films that are homogeneous and mechanically strong (tensile strength of 1-1.5 MPa). The modified Aquacoat® ECD was amenable to the Wurster coating process and can be coated on to drug layered sugar-pellets. 18

Dissolution studies evaluating drug release form the coated pellets showed that linear release profiles were possible if the excipient concentration and weight of the coating membrane could be optimized. The results also showed TEC was more effective than TA as a plasticizer and a pore-former, as linear drug-release profiles could be observed at lower concentrations and % weight gain for TEC-containing films.

REFERENCES: 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13.

14.

Kumari, M.H., Samatha, K., Balaji, A. & Shankar, M.U. RECENT NOVEL ADVANDCEMENTS IN PELLET FORMULATION: A REVIEW. International Journal of Pharmaceutical Sciences and Research 4, 3803 (2013). Dey, N., Majumdar, S. & Rao, M. Multiparticulate drug delivery systems for controlled release. Tropical Journal of Pharmaceutical Research 7, 1067-1075 (2008). Rhee, Y.-S., Lee, J.-H., Lee, B.-J. & Park, E.-S. Controlled-release pelletized dosage forms using the extrusion-spheronization process. J Pharm Investigation 40, 103-112 (2010). Ibrahim, M.A. Formulation and evaluation of mefenamic acid sustained release matrix pellets. Acta pharmaceutica 63, 85-98 (2013). Mehta, K.A., Kislalioglu, M.S., Phuapradit, W., Malick, A.W. & Shah, N.H. Release performance of a poorly soluble drug from a novel, Eudragit-based multi-unit erosion matrix. Int J Pharm 213, 712 (2001). Gautam, N. & Trivedi, P. FORMULATION & DEVELOPMENT OF PELLETS OF TOLTERODINE TARTRATE: A QUALITATIVE STUDY ON WURSTER BASED FLUIDIZED BED COATING TECHNOLOGY. Journal of Drug Delivery and Therapeutics 2 (2012). Villa Nova, M. et al. Formulation and characterization of ethylcellulose microparticles containing. l-alanyl-l-glutamine peptide. Drug development and industrial pharmacy 40, 13081317 (2014). Garekani, H.A., Dolatabadi, R., Akhgari, A., Abbaspour, M.R. & Sadeghi, F. Evaluation of ethylcellulose and its pseudolatex (Surelease) in preparation of matrix pellets of theophylline using extrusion-spheronization. Iranian journal of basic medical sciences 20, 9-16 (2017). Sánchez-Lafuente, C. et al. Development of sustained release matrix tablets of didanosine containing methacrylic and ethylcellulose polymers. International journal of pharmaceutics 234, 213-221 (2002). Shin, T.H. et al. Formulation and in vivo pharmacokinetic evaluation of ethyl cellulose-coated sustained release multiple-unit system of tacrolimus. International journal of biological macromolecules 109, 544-550 (2018). Kazlauske, J. et al. Determination of the release mechanism of Theophylline from pellets coated with Surelease((R))-A water dispersion of ethyl cellulose. Int J Pharm 528, 345-353 (2017). Gong, W. et al. Preparation, Characterization and in vivo Evaluation of Simple Monolithic Ethylcellulose-coated Pellets Containing Topiramate with Biphasic Release Characteristics. Current drug delivery 13, 245-254 (2016). Wesseling, M. & Bodmeier, R. Drug release from beads coated with an aqueous colloidal ethylcellulose dispersion, Aquacoat, or an organic ethylcellulose solution. European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V 47, 33-38 (1999). Carlin, B., Li, J. & Felton, L.A. Pseudolatex dispersions for controlled drug delivery. DRUGS AND THE PHARMACEUTICAL SCIENCES 176, 1 (2008).

19

15. 16. 17.

18.

19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32.

Miller, D.A. & McGinity, J.W. Aqueous polymeric film coating. Pharmaceutical dosage forms: Tablets 1, 399-437 (2008). Bodmeier, R. & Paeratakul, O. The distribution of plasticizers between aqueous and polymer phases in aqueous colloidal polymer dispersions. International journal of Pharmaceutics 103, 4754 (1994). Frohoff-Hülsmann, M.A., Schmitz, A. & Lippold, B.C. Aqueous ethyl cellulose dispersions containing plasticizers of different water solubility and hydroxypropyl methylcellulose as coating material for diffusion pellets: I. Drug release rates from coated pellets. International Journal of Pharmaceutics 177, 69-82 (1999). Frohoff-Hulsmann, M.A., Lippold, B.C. & McGinity, J.W. Aqueous ethyl cellulose dispersion containing plasticizers of different water solubility and hydroxypropyl methyl-cellulose as coating material for diffusion pellets II: properties of sprayed films. European journal of pharmaceutics and biopharmaceutics : official journal of Arbeitsgemeinschaft fur Pharmazeutische Verfahrenstechnik e.V 48, 67-75 (1999). Siepmann, J., Siepmann, F., Paeratakul, O. & Bodmeier, R. Process and formulation factors affecting drug release from pellets coated with Ethylcellulose pseudolatex Aquacoat®. DRUGS AND THE PHARMACEUTICAL SCIENCES 176, 203 (2008). Siepmann, F., Hoffmann, A., Leclercq, B., Carlin, B. & Siepmann, J. How to adjust desired drug release patterns from ethylcellulose-coated dosage forms. Journal of Controlled Release 119, 182-189 (2007). Gunder, W., Lippold, B. & Lippold, B. Release of drugs from ethyl cellulose microcapsules (diffusion pellets) with pore formers and pore fusion. European journal of pharmaceutical sciences 3, 203-214 (1995). Ghebre-Sellassie, I. & Harris, M.R. in Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms, Third Edition 67-86 (CRC Press, 2008). Wong, D. & Bodmeier, R. Flocculation of an aqueous colloidal ethyl cellulose dispersion (Aquacoat®) with a water-soluble polymer, hydroxypropyl methylcellulose. European journal of pharmaceutics and biopharmaceutics 42, 12-15 (1996). Sakellariou, P. & Rowe, R.C. The morphology of blends of ethylcellulose with hydroxypropyl methylcellulose as used in film coating. International Journal of Pharmaceutics 125, 289-296 (1995). Dashevsky, A. et al. Effect of water-soluble polymers on the physical stability of aqueous polymeric dispersions and their implications on the drug release from coated pellets. Drug Dev Ind Pharm 36, 152-160 (2010). Yue, Y., Collaku, A. & Liu, D.J. Evaluation of a 12-Hour Sustained-Release Acetaminophen (Paracetamol) Formulation: A Randomized, 3-Way Crossover Pharmacokinetic and Safety Study in Healthy Volunteers. Clinical pharmacology in drug development 7, 95-101 (2018). Pham, L. & Christensen, J.M. Preparation of acetaminophen capsules containing beads prepared by hot-melt direct blend coating. Pharm Dev Technol 19, 91-102 (2014). Parikh, N.H., Porter, S.C. & Rohera, B.D. Tensile properties of free films cast from aqueous ethylcellulose dispersions. Pharmaceutical research 10, 810-815 (1993). Rohera, B.D. & Parikh, N.H. Influence of type and level of water-soluble additives on drug release and surface and mechanical properties of Surelease® films. Pharmaceutical development and technology 7, 421-432 (2002). Felton, L.A., O’Donnell, P.B. & McGinity, J.W. in Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms, Third Edition 125-148 (CRC Press, 2008). PubChem Triethyl Citrate. PubChem Triacetin. 20

33.

Bodmeier, R. & Paeratakul, O. Plasticizer uptake by aqueous colloidal polymer dispersions used for the coating of solid dosage forms. International journal of pharmaceutics 152, 17-26 (1997).

21

Graphical abstract

22

Conflict of interest No conflict of interest exists for the authors or co-authors with regards to the data presented in this manuscript.

23