- Email: [email protected]
A mechanistic understanding of compression damage to the dissolubility of coated pellets in tablets
Journal Pre-proofs Research paper A Mechanistic Understanding of Compression Damage to the Dissolubility of Coated Pellets in Tablets Tze Ning Hiew, Yu Harn Tian, Huei Ming Tan, Paul Wan Sia Heng PII: DOI: Reference:
S0939-6411(19)31305-0 https://doi.org/10.1016/j.ejpb.2019.11.006 EJPB 13182
To appear in:
European Journal of Pharmaceutics and Biopharmaceutics
Received Date: Revised Date: Accepted Date:
26 August 2019 17 November 2019 25 November 2019
Please cite this article as: T.N. Hiew, Y.H. Tian, H.M. Tan, P.W.S. Heng, A Mechanistic Understanding of Compression Damage to the Dissolubility of Coated Pellets in Tablets, European Journal of Pharmaceutics and Biopharmaceutics (2019), doi: https://doi.org/10.1016/j.ejpb.2019.11.006
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.
A Mechanistic Understanding of Compression Damage to the Dissolubility of Coated Pellets in Tablets
Tze Ning Hiewa, Yu Harn Tiana, Huei Ming Tanb, Paul Wan Sia Henga,*
aGEA-NUS
Pharmaceutical Processing Research Laboratory, Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore 117543, Singapore.
bEngineering
Science Programme, Faculty of Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117575, Singapore.
*Corresponding author: P.W. S. Heng. Tel.: +65 65162930; fax: +65 67752265. Email: [email protected]
1
Abstract Damage to the drug diffusion coat barrier of controlled release pellets by the compaction force when preparing multiple-unit pellet system tablets is a major concern. Previous studies have shown that pellets located at the tablet axial and radial peripheral surfaces were more susceptible to damage when compacted due to the considerable shear encountered at these locations. Hence, this study was designed to assess with precision the impact of pellet spatial position in the compact on the extent of coat damage by the compaction force via a single pellet in minitablet (SPIM) system. Microcrystalline cellulose (MCC) pellet cores were consecutively coated with a drug layer followed by a sustained-release layer. Chlorpheniramine maleate was the model drug used. Using a compaction simulator, the coated pellets were compacted singly into 3 mm diameter SPIMs with MCC as the filler. SPIMs with individual pellets placed in seven positions were prepared. The uncompacted and compacted coated pellets, as SPIMs, were subjected to drug release testing. The dissolution results showed that pellets placed at the topradial position were the most susceptible to coat damage by the compaction force, while pellets positioned within the minitablet at the middle and upper quadrant positions showed the least damage. The SPIM system was found to be effective at defining the extent of coat damage to the pellet spatial position in the compact. This study confirmed that coated pellets located at the periphery were more susceptible to damage by compaction, with pellets located at the top-radial position showing the greatest extent of coat damage. However, if the pellet was completely encrusted by the cushioning filler, coat damage could be mitigated. Further investigations were directed at how the extent of coat damage impacted drug release. Interestingly, small punctures were found to be most detrimental to drug release whilst coats with large surface cuts did not completely fail. A damaged pellet coat has some self-sealing ability and failure is not total. Thus, this study provides a deeper understanding of the consequence of coat damage to drug release when sustained release coated pellets are breached.
Keywords: coated pellet, minitablet, microcrystalline cellulose, sustained release, compaction, coat damage
2
1.
Introduction
Controlled release dosage forms are very popular with patients as these dosage forms offer the advantages of reduced pill burden and dosing frequency, which could potentially manage patient noncompliance more effectively. Regulating the rate of drug release also impacts therapeutic efficacy as a gradual drug release in the gastrointestinal tract over a prolonged period provides better pharmacological outcomes. Multi-particulate dosage forms, commonly comprising of coated and uncoated pellets are advantageous as after administration, the pellets distribute uniformly in the gastrointestinal tract without high local concentration and are not prone to catastrophic dose dumping should some coated pellets fail. The key to developing multi-particulate dosage forms is the inclusion of slow release coated pellets. Pellet-based dosage forms are often delivered in gelatin capsules which can be contentious due to the use of animal protein. The solution is therefore the introduction of compacted pellets as multi-unit pellet system (MUPS) tablets. However, a common concern associated with MUPS tableting is the preservation of the integrity of the functional coat often used to modify drug release [1]. The film coatings used to modify drug release are usually acrylic- or cellulosic-based [2], with polymethacrylate and ethyl cellulose as the most commonly used polymers, respectively. The mechanical properties of the polymeric film are important for determining its durability and the ability to withstand the applied pressure during tableting. Due to their flexibility, acrylic polymers were reported to form more deformable films on pellets meant for compaction into MUPS tablets, whereas ethyl cellulose was more brittle by comparison [3].
In addition to the choice of coating polymer used, excipients as fillers are very important in MUPS tableting. Ideally, the filler should be able to cushion the coated pellets from compression force related damage and yet, make mechanically strong tablets which can readily disintegrate upon ingestion. The challenges presented in designing MUPS tablets require substantial research work at understanding how cushioning fillers can mitigate compaction-induced pellet coat damage and the conditions which may potentially aggravate or ameliorate the extent of damage. In studies using lactose, the particle size of the lactose was a critical attribute and micronized lactose was reported to be best [4, 5]. Highly
3
compressible excipients were also found to be good cushioning fillers [6-9]. Besides blending pellets with cushioning excipients, attempts were also made to directly deposit the cushioning excipients as an overcoat on the coated pellets [10-13].
The volumetric ratio between the filler and pellets or, critical pellet volume fraction [14] was reported to influence the extent of pellet coat damage. Critical pellet volume fraction values ranging from 0.374 [14] to 0.39 [4] were optimal to maintain a percolating network of cushioning fillers around the pellets, thereby minimizing pellet-pellet contacts during compression. In addition, the ratio between cushioning filler and pellets was also found to influence mechanical strength, friability and disintegration time of the resultant tablets [15].
While the coating polymer, cushioning filler and pellet volume fraction are critical attributes to successful MUPS tablet design, one area in MUPS tableting which can potentially impact release performance of the coated pellets but not well investigated or understood is the effect of pellet spatial position on coat damage. A previous study excavated pellets from different locations of the MUPS tablets and it was reported that the spatial location of a pellet within the tablet played an important role at determining the extent of compaction-related pellet coat damage [14]. Pellets located at the peripheral surfaces, be they axial or radial surfaces, had pellet coats more deformed or damaged, especially when compression pressure used was above the crushing strength of the pellets.
Herein, the effect of pellet spatial position in a compact on coat damage due to compression was investigated using a single pellet in minitablet (SPIM) system. With the SPIM system, pellets, singly or a few together, can be accurately placed in the desired position of the compact and spatial-related factors such as potential damage due to pellet-pellet or pellet-wall contacts and pellet-filler interactions could be better understood. In this study, pellets were placed singly at seven different positions and compacted into minitablets with microcrystalline cellulose as the cushioning filler. For SPIM pellets in contact with the tool surfaces, they were on top or bottom in contact with the punches and mid-radial or top-
4
radial in contact with the die wall. For pellets located centrally inside the tablet, they were placed at either the upper quadrant, middle or lower quadrant. Dissolution studies were conducted on the compacted pellets to determine the extent of pellet coat damage.
5
2. 2.1.
Materials and methods Materials
Chlorpheniramine maleate (CPM; BP grade, China) was used as the model drug in this study. Hydroxypropyl methylcellulose (HPMC; Methocel VLV, Dow Chemical, USA) was used as supplied for drug layering. Aqueous poly(meth)acrylates dispersion (ERS; Eudragit RS30D, Evonik, Germany) was used as the sustained-release coating polymer, with triethyl citrate (TEC; Merck, Germany) as the plasticizer. Talc (Chemipure, Singapore) was milled in a pin mill (ZM1000, Retsch, Germany) and used as an anti-tack agent in the sustained-release film coating. Microcrystalline cellulose (MCC) cores (500710 μm; Celphere CP-507, Asahi Kasei, Japan) were the starter cores for drug layering, followed by sustained release coating. Degassed purified water was the dissolution medium. MCC (Ceolus PH-101, Asahi Kasei, Japan) was the filler cushioning agent in the SPIM with sodium starch glycolate (SSG; DFE Pharma, Germany) as the disintegrant. Potassium dihydrogen phosphate (KH2PO4; Merck, Germany), ortho-phosphoric acid (H3PO4; Merck, Germany) and methanol (HPLC grade, Fisher Chemical, Canada) were used to prepare the mobile phase for high performance liquid chromatography (HPLC).
2.2.
Pellet coating
The MCC cores were consecutively coated with a drug layer followed by the sustained release layer in an air suspension coater (FlexStream module, MP-1, GEA, UK). For drug loading, 1.8 kg of MCC cores were layered with a coating media containing 7%, w/w CPM and 14%, w/w VLV HPMC to a final coat weight gain of 21%, w/w, calculated based on the percentage ratio of dry weight of coat material deposited onto the weight of the cores. The pellets were then oven dried overnight at 60 °C. The pellets were then fractionated using 500 and 1000 μm aperture size sieves (Endecotts, UK) to remove the fines and agglomerates, respectively.
The drug loaded 500-1000 μm pellet fraction was overcoated with poly(meth)acrylate polymer. The coating media consisted of 39.2%, w/w ERS, 2.4%, w/w TEC and 5.9%, w/w talc, which corresponded
6
to a final dry coat composition of 58.9%, w/w polymer, 11.7%, w/w TEC and 29.4%, w/w talc. The pellets were coated to 17.5%, w/w coat weight gain. Upon completion, the coated pellets were cured in-process for 30 min with water, and further cured for 12 h at 40 °C in a hot air oven. The pellets were then again fractionated to 500-1000 μm. The conditions and process parameters used for the coating of both the drug and sustained-release layers, optimized prior to this study, are shown in Table 1.
2.3. 2.3.1.
Physical characterization of pellets Pellet size and shape determination
Randomly selected pellets, about 200, were imaged (DP71, Olympus, Japan) using an optical microscope (SZ61, Olympus, Japan) and the captured images of 200 randomly selected pellets captured were analyzed (Image-Pro 6.3, Media Cybernatics, USA) for pellet mean Feret’s diameter, defined as the average caliper length of the pellet. Additionally, two shape determinant parameters, roundness and aspect ratio, were also calculated using Eqs. (1) and (2), respectively. Roundness reflects the sphericity of the pellets, while aspect ratio represents the extent of pellet elongation [16]. Therefore, the roundness and aspect ratio of a perfectly spherical pellet would be unity. 𝑝2 𝑅𝑜𝑢𝑛𝑑𝑛𝑒𝑠𝑠 = 4 × 𝜋 ×𝐴
𝐴𝑠𝑝𝑒𝑐𝑡 𝑅𝑎𝑡𝑖𝑜 =
(1)
𝑙 𝑏
(2)
where p is the perimeter, A is the area of the pellet, l and b are the major and minor axes of the ellipse equivalent to the particle, respectively.
2.3.2.
Mechanical testing of pellets
The crushing force of 20 pellets was determined by diametrical compression between two platens on a tensile tester (EZ Test, Shimadzu, Japan) fitted with a 100 N load cell and averaged. The crushing rate used was 2 mm/min, and the maximum load (Fm) required to crush a pellet was recorded. The tensile (crushing) strength of each pellet was calculated using Eq. (3):
7
𝐶𝑟𝑢𝑠ℎ𝑖𝑛𝑔 𝑠𝑡𝑟𝑒𝑛𝑔𝑡ℎ =
𝐹m 2
𝜋𝑑
(3)
×4
where d is the mean Feret’s diameter of the pellets, determined via optical microscopy.
2.4. 2.4.1.
Preparation of SPIM Optimization of compression pressure and filler composition
Powder blends of MCC and 0.5 or 1%, w/w SSG were compacted into 3 mm minitablets with a compaction simulator (STYL’One, Medelpharm, France) fitted with multiple-tip, standard concave tooling (8 tips per punch; Natoli Engineering Company, US). Control minitablets of only MCC were also prepared. A fill depth of 6 mm (1.5 × tablet diameter), which produced approximately 12 mg minitablets, was used. The minitablets were compacted to five pressures, per punch tip, were 34, 68, 102, 136 and 170 MPa. Disintegration tests (The United States Pharmacopeia method; DT2, Sotax, Switzerland) were carried out at 37 ± 2 °C in purified water with at least six randomly chosen minitablets and the results were averaged.
2.4.2.
Compression of SPIM on compaction simulator
2.4.2.1. Design of pellet placement guide for SPIM preparation A pellet placement guide (Figure 1) was designed to accurately insert a pellet centrally within the die cavity. Using an aluminum (6061 alloy) disc of 22.5 mm diameter and 7 mm thick, eight protrusions were milled, each with 1.0 mm through holes drilled. The design and fabrication work was accomplished with the use of 3D computer-aided design software (SOLIDWORKS 2018 SP02, Dassault Systèmes, France), computer-aided manufacturing software (HSMWorks Ultimate 2019, Autodesk, US), 3D desktop Computer Numerical Control machine (Nomad 883 Pro, Carbide 3D, US), 1/8” flat, ¼” corner rounding, 30° and 60° carbide end mills, and a 1.0 mm HSS drill bit.
8
2.4.2.2. Spatial pellet placements in SPIMs The coated pellets were compacted singly at 65 MPa using the equal force mode. Each SPIM contained approximately 12 mg of MCC pre-blended with 1%, w/w SSG as the disintegrant. In each SPIM, a pellet was placed at one of the seven pre-selected positions. For tablet surface placement, pellet was deposited at either the top, bottom, mid-radial or top-radial positions. For pellet centrally positioned inside the SPIM, it was placed using the placement guide at the lower quadrant, middle or upper quadrant of the tablet, assisted by bi-layer filling with pre-loading of 25%, 50% or 75%, respectively, of the total feed weight of the intended tablet feed, followed by the pellet and then the balance of the tablet weight (Figure 1).
Preparation of SPIM with pellet at the bottom was carried out by carefully placing a pellet centrally on the concavity of the power punch, followed by 12 mg of the pre-weighed blended filler. Conversely, SPIM with pellet at the top was prepared by first filling all the filler followed by careful placement of pellet centrally at the top with the aid of a placement guide.
For SPIMs containing pellets at the mid-radial and top-radial positions, an additional tamping step was added prior to pellet placement to ensure that the pellets could be placed firmly at the circumferential side of the die cavity, in contact with the die wall.
2.5. 2.5.1.
Drug release testing Optimization of conditions for drug release testing
A coloring agent (Eurolake Sunset Yellow, AJ4929, Netherlands) was hand-blended with MCC at a 1:1 gravimetric ratio for 2 min. The powder blend was then compacted into 100 mg compacts of 6 mm diameter using a single-station press (NP-RD10, Natoli Engineering Company, US) installed with standard concave tooling (Natoli Engineering Company, US). These compacts were then placed individually into 1.4 cm diameter test tubes containing 8 mL of deionized water maintained at 37 ± 0.5 °C by a thermostatically controlled shaker water bath. The shaker water bath was oscillated at 40, 60 or
9
80 oscillations per min, with regular visual inspection for homogeneity of the color distribution in the test tubes. After 15 min, 2 mL of the medium was withdrawn and replaced with an equal volume of deionized water, and the color distribution monitored for another 15 min. Based on visual inspection, 40 oscillations per min was found to be ideal and used for the subsequent drug release experiments.
2.5.2.
Drug release from individual pellets
The uncompacted and compacted coated pellets, as SPIMs, were subjected to drug release testing in deionized water. Each dosage form was placed in a test tube of 1.4 cm diameter containing 8 mL of deionized water maintained at 37 ± 0.5 °C in a shaker water bath pendulating at 40 oscillations per min. At time points of 15, 30, 60 and 120 min, 2 mL aliquots were withdrawn using a mechanical pipette and replaced with fresh media. After 120 min, the dosage form was crushed with a glass rod and the remnants left to shake for another 15 min to allow for complete dissolution of the drug, after which 2 mL of dissolution medium was withdrawn. Dissolution tests were performed with at least three randomly chosen pellets/tablets and the results were averaged. The drug content in each sample was determined using HPLC.
From the plot of cumulative drug release against time, T50%, the time required for 50% drug release, was determined. Additionally, similarity factor f2, which was calculated based on Eq.(4), was used to determine if the release profiles from uncompacted pellets and compacted pellets were similar. f2 is a logarithmic reciprocal square root transformation of the sum of squared error and is a measurement of the similarity in the percent (%) dissolution between the two curves. A f2 value between 50 and 100 indicates equivalence [17]. 1 𝑓2 = 50 × log {[1 + ( ) 𝑛
2.5.3.
𝑛
∑
|𝑅𝑖 ― 𝑇𝑖|2 ] ―0.5 × 100}
(4)
𝑖=1
High performance liquid chromatography
Drug contents were determined using a HPLC system (LC-2010C HT, Shimadzu, Japan) with a mobile phase of 50%, v/v phosphate buffer (H3PO4/KH2PO4, pH 3.0, 50 mM) and 50%, v/v methanol delivered, 10
isocratic, at 1 mL/min through a 100 mm × 4.6 mm C18 column (ACE Generix, 5 μm, Advanced Chromatography Technologies, UK) maintained at 40 °C. A wavelength of 260 nm was employed for detecting CPM at its retention time of 3.25 min.
2.6.
Preparation of pellets with different cut depth
Deep transverse cuts were made to randomly selected pellets to simulate compaction damage on pellet coatings. Reproducible linear cuts were enabled by accurately drilled seat holes of various depths on an aluminum plate. The fabricated plate is a 100 mm × 50 mm × 3 mm thick 6061 aluminum alloy plate, drilled with 1.0 mm diameter holes, of various depths from 0.2-1.0 mm, in 0.1 mm increments in depths, and hole pitch of 10 mm. Design and fabrication was carried out as per pellet placement guide (Section 2.4.2.1). The drilling process was done using a micro solid end mill (905L010-MEGA-T, SECO Tools, Sweden), programmed with 10,000 rpm spindle speed and 20 mm/s plunge feed rate.
Each test pellet was inserted in a hole of target cut depth and vertically bisected from the top with the use of a sharp surgical blade until the blade edge was flush with the surface of the aluminum plate. Selected pellets were also completely bisected to mimic a completely broken pellet. At least five pellets were prepared for each cut depth, and drug release rates determined.
11
3. 3.1.
Results Physical characterization of pellets
The Feret’s diameter, roundness and aspect ratio of the coated pellets were 735.0 ± 42.3 μm, 1.14 ± 0.08 and 1.10 ± 0.07, respectively, from pellet images and by image analysis. Using diametrical compression, the crushing strength of the pellets was found to be 33.6 ± 4.3 MPa.
3.2.
Disintegration time of minitablets
The disintegration time of the SPIM was expected to influence the rate of drug release, as nondisintegrating SPIMs could potentially confound drug dissolution from the coated pellets. SPIMs containing pellets may be encrusted by the filler and are susceptible to lower or no drug release despite being in contact with the dissolution medium. Non-disintegrating minitablets is especially a concern with MCC as it was previously reported to exert a shielding effect that slowed down drug release [18].
As shown in Figure 2, disintegration time of the minitablets increased with compression pressure. Without disintegrant aid, they failed to disintegrate even after 30 min when compacted at 170 MPa. An addition of SSG enabled all the minitablets to disintegrate within 30 s. The contribution of SSG to reduce disintegration time was especially marked at high compression pressures of 136 MPa and 170 MPa. For minitablets compacted at 136 MPa, disintegration time shortened to 20.4 s and 12.6 s when 0.5%, w/w and 1%, w/w SSG were added, respectively. Hence, MCC with 1%, w/w SSG was chosen for the formulation of SPIM.
3.3.
Effect of pellet position on drug release
Dissolution studies on the uncompacted and compacted pellets, as SPIM, were carried out to elicit the impact of pellet position in the compact on its rate of drug release. As the pellets were coated with a sustained-release polymer, any compromise to coat integrity would be expected to result in faster drug release. Drug release profiles of the pellets over 120 min, both uncompacted and compacted, are shown in Figure 3a.
12
Regardless of the position of the pellet in the SPIM, all the compacted pellets showed faster drug release compared to the uncompacted pellets, reflecting some damage to the sustained release coat caused by the compaction force. This was further substantiated by the f2 values (Table 2a), which showed that with the exception of pellets placed at the upper quadrant position, the drug release profiles of the other compacted pellets were significantly different from that of the uncompacted pellets. However, differences in the rate of drug release were observed for the different SPIMs, suggesting that pellets placed at different positions were damaged to different extents. Pellets placed at the middle and upper quadrant positions showed the least damage. There was increased severity of damage but to rather similar extents for pellets placed at the top, bottom, lower quadrant and mid-radial positions. Pellets placed at the top-radial position was most damaged, as reflected by fastest drug release rate and dissolution T50% (Figure 3b) averaging 37.3 min, or about 2.5 times shorter than that of the uncompacted pellets (92.8 min).
3.4.
Effect of pellet cut depth on drug release
Observation from the dissolution studies of the coated pellets placed at different positions within the SPIM appeared to show different degrees of coat damage but unexpectedly, not a total failure. As shown in Figure 3a, drug release profiles can broadly be sub-divided into 3 clusters. In order to better understand the findings of this study, coated pellets with different cut depths were prepared to mimic the severity of coat damage by the extent of crack formation caused by compression during tablet making. By depositing the coated pellets in pre-drilled 1 mm diameter pits of depths, 0.2 mm, 0.3 mm and 0.4 mm, circumferential slits were made on the pellets using a sharp razor, bisecting the exposed portion. By computation based on the mean Feret’s diameter of the pellets and the depth of the hole, the cuts could be translated to approximately 65, 56 and 47% cracks along the circumference of the pellets placed in 0.2 mm, 0.3 mm and 0.4 mm pits, respectively.
Figure 4b shows the dissolution profiles of the cut pellets. Pellets that were bisected showed the fastest drug release, with a T50% of 16.9 min, and 90% of the drug was released within an hour. This was 13
followed by deepest cut coated pellets in 0.2 mm depth pit, which had a T50% of 18.8 min, close to that of the fully bisected pellets. Pellets cut in 0.3 mm deep pit had a longer T50% of 47 min. Interestingly, pellets cut in the 0.4 mm deep pit, with the smallest fraction of coat damage, had a rather different drug release profile, showing a sigmoidal pattern. The amount of drug released from these pellets during the first 15 min was comparable to that of the uncompacted pellets, suggesting that the puncture did not caused any immediate change in release rate but a latent burst release occurred which pointed to a rather catastrophic consequence of coat failure, as the later high release rate paralleled that of a fully bisected coated pellet.
14
4.
Discussion
Fabrication of tablets invariably involves the application of compression pressure to a powder bed to consolidate the particles placed within a confined space into a coherent mass of fixed geometry [19]. When the pressure is applied, the particles first undergo rearrangement which results in further volume reduction and an increase in compact porosity. When the interparticulate frictional forces are too high for further volume reduction, the particles will undergo deformation [19]. Studies have shown that the distribution of compact density and mechanical properties across the matrix of a tablet are unequal. In flat face tablets, despite equal force being applied over the powder surface by the upper and lower punches, variability had been reported [20, 21]. Therefore, it is not unexpected that the forces experienced during fabrication and resultant density distribution in a biconvex tablet is nonhomogeneous. Density distribution analyses of biconvex tablets showed that the tablet density was in general the lowest around the center of the tablet, and gradually increased laterally towards the edge of the tablet, where density was the highest [22-26]. This was attributed to the localized densification of the powder as the biconvex shape resulted in thinner volume along the peripheral edge. This nonhomogeneity in density distribution will undoubtedly occur within a MUPS tablet and potentially be translated to locality-dependent differences in the forces exerted on the pellets during compaction. Indeed, pellets located at both the radial and axial surfaces within MUPS tablets were found to suffer a larger extent of deformation and coat damage due to direct impact and friction with the compression tools [14]. Building on the reported observations, this study employed the use of SPIM to more precisely examine the extent of pellet coat damage in relation to the spatial position of the pellets and the ultimate consequence to the damaged pellets in their performance as sustained drug release vehicles.
The dissolution results from the SPIMs showed that pellet spatial position played a definitive role in the extent of coat damage. Pellets placed at the top-radial position was most damaged, as evident from the shortest dissolution T50% as compared to pellets in other locations. As biconvex minitablets were made, the pellets at the top-radial position received the highest impact forces from compression pressure exerted along with friction against the compaction tools, which collectively inflict considerable damage to the sustained release coat. The other pellets in contact to tablet surfaces, at the top, bottom and mid15
radial positions, were also damaged but slightly less than pellets at the top-radial position. Nonetheless, the drug release profiles suggest rather severe damage to the functional coat, impairing somewhat their diffusion barrier function. Interestingly, pellets located at the lower quadrant of the SPIM and encased by the filler material had rather similar release profiles as those located at the top, bottom and midradial positions. In contrast, pellets located at the middle and upper quadrant positions were least damaged and had similar drug release profiles. As the SPIMs were compacted by simultaneously moving the upper and lower punches in opposing direction, the forces exerted biaxially should theoretically be close. The consequence of this mechanical operation was two-fold. Firstly, the compression pressure exerted by the lower punch was more sustained compared to the upper punch, as although the latter moved rapidly downwards, it was immediately withdrawn once the set thickness or maximum compression pressure was achieved. With the recorded dwell time ranging from 1-5 ms, the maximum compression pressure event was very transient. Secondly, examination of the upper and lower punch pressures revealed a disparity between them. From the compaction data of the SPIMs, the pressures exerted by the lower punch were consistently higher, by 8-16% higher than that of the upper punch. This observation would explain the higher degree of coat damage encountered by the pellets in the lower quadrant of the SPIMs as compared to those in the middle or upper quadrant.
The compaction force used to make the SPIMs in this study was 3.8 kN. With a cup area of 7.35 mm2, the compression pressure was 65 MPa, which was approximately twice the average crushing force of the pellets when they were crushed between two platens. This compression pressure was chosen to be equivalent to a 10 kN force applied on the powder bed with 14 mm flat faced tooling. Even at a rather high compression pressure applied, it was surprising to discover that the compacted coated pellets did not fail completely, and appearing to show gradated degrees of failure. It was originally assumed that a burst rapid release would result from a fractured pellet coat, and emulating the release profile of a bisected coated pellet, with more than 90% drug released within an hour. However, as a gradated degree of failure was observed for the compacted pellets, it could be hypothesized that some re-sealing mechanism existed, possibly by the hydrated polymeric gel layer underneath the sustained release coat. A previously published study examined the effect of pellet coat rupture on drug release by piercing the 16
pellet coat with a needle. Microscopic observation revealed a shift in the pellet core to an eccentric position brought about by the pressure build-up within the pellet, especially in the region away from the puncture point. The movement of the core material consisting of a hydrated polymeric gel and insoluble MCC core sealed off the puncture point, thereby trapping the drug within the pellet [27]. In addition, the swelling HPMC polymer also aided in re-sealing the punctured coat, thereby serving as an additional diffusion barrier for the drug.
While pellets with a small puncture exhibited self-sealing and preserving abilities, it was not known if this phenomenon would prevail with respect to compaction created cracks, often linear. The outcome from the dissolution studies showed that even with the cut coats, the pellets exhibited some cut depthdependent delay in drug release. Despite deep cuts made on the coats of the pellets, complete failure of the dosage form akin to a dose dump was not seen. However, the f2 values suggested that the drug release profiles were still significantly different from the uncompacted pellets (Table 2b). This finding shows some innate ability of the HPMC polymer to re-seal the breached sustained release coat. HPMC is known to swell when wetted, particularly with higher viscosity grades used in sustained release matrix tablets. HPMC may also erode and dissolve [28]. Therefore, the ability of HPMC to control and sustain drug release requires a fine balance between its rate of gel formation, swelling and erosion. In this study, the drug was dissolved in the VLV HPMC solution before it was coated onto the MCC cores. As the HPMC used for drug loading was of a very low viscosity grade, it is postulated that any gel formed could not possess sufficient gel strength to effectively seal the crack in the coat. Regardless, since the rate of drug release showed some correlation to the extent of coat damage, HPMC certainly had played a role in limiting drug release during dissolution.
Another hypothesis to explain the results is that in breached sustained release coat, the two breached edges could realign themselves and continue to act as a continuous diffusion barrier. For a typical coated pellet, drug release could proceed by the drug concentration gradient built-up across the barrier, from the internal drug reservoir to the bulk medium [29]. Thus, a tear to the diffusion coat may not be as
17
detrimental to its barrier properties, provided no material leakage or differential pressure build up to disrupt the status quo.
Eudragit RS30D contains water-insoluble copolymers of ethyl acrylate, methyl methacrylate and a low content of a methacrylic acid ester with quaternary ammonium groups (trimethylammonioethyl methacrylate chloride). The ammonium groups are present as salts and make the polymer coat permeable through pore formation when wetted. This allows water permeation into the pellet to dissolve and leach out the drug by diffusion (Figure 5-I). The acrylic diffusional barrier enables the formation of a saturated drug reservoir of dissolved CPM within the pellet, with constant drug diffusion across the water-insoluble, semi-permeable polymer coat. The reservoir of CPM within the pellet maintains a constant concentration gradient across the diffusional barrier. In the case of cut coats, faster drug release was observed due to some content leakage (Figure 5-III).
Interestingly, pellets with the shallower cuts exhibited a greater degree of burst release. This is counterintuitive as the pellets with the least damage should theoretically leverage most on the re-sealing ability of the breached coats to retard drug release. It is postulated that HPMC swelled when wetted and had contributed to the immediate re-sealing of the slit. However, with time, the osmotic pressure buildup occurred due to the presence of HPMC in the coated pellets. With an intact coat, there is isotropic osmotic pressure build-up, thus resulting in uniform swelling. However, with partially damaged pellet coat, there is anisotropic osmotic pressure build-up, with higher pressure accumulated away from the crack, which results in uneven swelling [27]. The internal pressure would push the core material content towards the breached opening at the cut of the coat. This effectively produced a weak osmotic pump system, propelling some pellet content out via the cut to equilibrate the osmotic pressure build-up within and release character is akin that of an osmotic pump system (Figure 5-II). With a larger crack, any osmotic pressure build-up would be less [30]. Thus, for the smallest cut, pellets cut in the 0.4 mm depth pit, had initial retarded drug release as shown by the lag phase but eventually, the osmotic pressure build-up was the release driver, which provided a steady rapid ejection of pellet liquid content along
18
with the drug. This drastic latent increase in drug release produced a profile much identical to that of the bisected pellets but the early lag phase was rather similar to that of the intact coated pellet.
By comparing the release profiles of the pellets from SPIMs to the pellets with different cut depths, it was observed the pellets placed at the top-radial position experienced a greater extent of coat damage compared to the other pellets, with the extent of damage similar to pellets cut in the 0.2 mm deep pit. The release profiles of the other pellets from SPIMs were more similar to the pellets cut in the 0.3 mm deep pit. The difference in drug release from pellets with different cut depths helped to further unravel the consequences of compaction-induced coat damage on the drug release profiles obtained from the compacted coated pellets as SPIMs. Clearly, pellet coat damage by compaction does not necessarily cause complete product failure as there could be differences in release rates as attributed to the different extents of breaches or cracks present. Even for pellets completely surrounded by the cushioning filler, their spatial position and extent of damage inflicted could be determinant of their sustained drug release function but may be in a rather complex manner.
19
5.
Conclusion
In this study, the SPIM system was devised and found beneficial for evaluating the effect of pellet spatial position on the extent of coat damage. Pellets located at the periphery were more susceptible to damage by compaction, with pellets located at the top-radial position showing the highest extent of coat damage. However, if the pellet was surrounded completely by the cushioning filler, the degree of coat damage could be mitigated. Additionally, the extent of pellet coat damage was also found to affect the re-sealing capability of the damaged coat. Therefore, even with the use of compression pressure above the crushing force of the pellets, the pellets did not fail completely in their sustained drug release function. In MUPS tablets, the pellets are randomly distributed within the tablet matrix and the ensemble drug release profile obtained represents a complex cacophony of release profiles by individual pellets located in the MUPS tablet. The role played by the pellet spatial position on pellet coat damage emphasizes the importance of formulating MUPS tablets below the critical pellet volume fraction, as this would reduce the number of pellets located at the surface of the tablets, thereby ensuring that the damage caused to the pellet coat could be kept to a minimum. The findings in this study provide important insights into the consequences of coat damage in MUPS tableting and clearly, a breached diffusion coat may have some degree of self-sealing properties.
20
Acknowledgements This research work was supported by the GEA-NUS PPRL fund (N-148-000-008-001). The authors also wish to acknowledge support from Asahi Kasei, Japan for the gratis supply of the Ceolus PH-101 and Celphere CP-507. The initial concept of SPIM arose from the project work supported by Asahi Kasei in MUPS tableting (R-148-000-275-597).
21
References [1] S.R. Béchard, J.C. Leroux, Coated pelletized dosage form: Effect of compaction on drug release, Drug Dev. Ind. Pharm., 18 (1992) 1927-1944. [2] L.A. Felton, P.B. O'Donnell, J.W. McGinity, Mechanical properties of polymeric films prepared from aqueous dispersions, in: J.W. McGinity, L.A. Felton (Eds.) Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms, CRC Press, Boca Raton, FL, 2008, pp. 105-128. [3] R. Bodmeier, O. Paeratakul, Mechanical properties of dry and wet cellulosic and acrylic films prepared from aqueous colloidal polymer dispersions used in the coating of solid dosage forms, Pharm. Res., 11 (1994) 882-888. [4] W.C. Chin, L.W. Chan, P.W.S. Heng, A mechanistic investigation on the utilization of lactose as a protective agent for multi-unit pellet systems, Pharm. Dev. Technol., 21 (2016) 222-230. [5] X. Lin, W.C. Chin, K.-f. Ruan, Y. Feng, P.W.S. Heng, Development of potential novel cushioning agents for the compaction of coated multi-particulates by co-processing micronized lactose with polymers, Eur. J. Pharm. Biopharm., 79 (2011) 406-415. [6] G.J. Vergote, F. Kiekens, C. Vervaet, J.P. Remon, Wax beads as cushioning agents during the compression of coated diltiazem pellets, Eur. J. Pharm. Sci., 17 (2002) 145-151. [7] F. Zeeshan, K.K. Peh, Y.T.F. Tan, Exploring the potential of a highly compressible microcrystalline cellulose as novel tabletting excipient in the compaction of extended-release coated pellets containing an extremely water-soluble model drug, AAPS PharmSciTech, 10 (2009) 850. [8] J.J. Torrado, L.L. Augsburger, Effect of different excipients on the tableting of coated particles, Int. J. Pharm., 106 (1994) 149-155. [9] S.U. Kucera, J.C. DiNunzio, N. Kaneko, J.W. McGinity, Evaluation of Ceolus™ microcrystalline cellulose grades for the direct compression of enteric-coated pellets, Drug Dev. Ind. Pharm., 38 (2012) 341-350. [10] X. Pan, M. Chen, K. Han, X. Peng, X. Wen, B. Chen, J. Wang, G. Li, C. Wu, Novel compaction techniques with pellet-containing granules, Eur. J. Pharm. Biopharm., 75 (2010) 436-442. [11] S.A. Altaf, S.W. Hoag, J.W. Ayres, Bead compacts. I. Effect of compression on maintenance of polymer coat integrity in multilayered bead formulations, Drug Dev. Ind. Pharm., 24 (1998) 737-746. [12] S.A. Altaf, S.W. Hoag, J.W. Ayres, Bead compacts. II. Evaluation of rapidly disintegrating nonsegregating compressed bead formulations, Drug Dev. Ind. Pharm., 25 (1999) 635-642. [13] A. Hosseini, M. Körber, R. Bodmeier, Direct compression of cushion-layered ethyl cellulosecoated extended release pellets into rapidly disintegrating tablets without changes in the release profile, Int. J. Pharm., 457 (2013) 503-509.
22
[14] M. Xu, P.W.S. Heng, C.V. Liew, Formulation and process strategies to minimize coat damage for compaction of coated pellets in a rotary tablet press: A mechanistic view, Int. J. Pharm., 499 (2016) 2937. [15] A. Debunne, C. Vervaet, D. Mangelings, J.-P. Remon, Compaction of enteric-coated pellets: influence of formulation and process parameters on tablet properties and in vivo evaluation, Eur. J. Pharm. Sci., 22 (2004) 305-314. [16] A.M. Bouwman, J.C. Bosma, P. Vonk, J.A. Wesselingh, H.W. Frijlink, Which shape factor(s) best describe granules?, Powder Technol., 146 (2004) 66-72. [17] Food and Drug Administration, Guidance for Industry: Dissolution Testing of Immediate Release Solid Oral Dosage Forms, Center for Drug Evaluation and Research (CDER), (1997). [18] T.N. Hiew, M.I.B. Alaudin, S.M. Chua, P.W.S. Heng, A study of the impact of excipient shielding on initial drug release using UV imaging, Int. J. Pharm., 553 (2018) 229-237. [19] G. Alderborn, Tablets and compaction, in: M.E. Aulton, K.M.G. Taylor (Eds.) Aulton's Pharmaceutics: The Design and Manufacture of Medicines, Elsevier, Edinburgh, 2013, pp. 504-549. [20] A. Kandeil, M.C. de Malherbe, S. Critchley, M. Dokainish, The use of hardness in the study of compaction behaviour and die loading, Powder Technol., 17 (1977) 253-257. [21] D. Sixsmith, D. McCluskey, The effect of punch tip geometry on powder movement during the tableting process, J. Pharm. Pharmacol., 33 (1981) 79-81. [22] I.C. Sinka, J.C. Cunningham, A. Zavaliangos, Analysis of tablet compaction. II. Finite element analysis of density distributions in convex tablets, J. Pharm. Sci., 93 (2004) 2040-2053. [23] Y. Hattori, R. Aoki, M. Otsuka, Use of partial least-squares analysis and fractionated X-ray computed tomography images in the investigation of density distribution of round tablets, Powder Technol., 302 (2016) 261-264. [24] R.K. May, K. Su, L. Han, S. Zhong, J.A. Elliott, L.F. Gladden, M. Evans, Y. Shen, J.A. Zeitler, Hardness and density distributions of pharmaceutical tablets measured by terahertz pulsed imaging, J. Pharm. Sci., 102 (2013) 2179-2186. [25] A. Djemai, I.C. Sinka, NMR imaging of density distributions in tablets, Int. J. Pharm., 319 (2006) 55-62. [26] B. Eiliazadeh, K. Pitt, B. Briscoe, Effects of punch geometry on powder movement during pharmaceutical tabletting processes, Int. J. Solids Struct., 41 (2004) 5967-5977. [27] P.W.S. Heng, L.W. Chan, S.H. Chew, Mechanism of pellet coat rupture and its effect on drug release, Chem. Pharm. Bull. (Tokyo), 47 (1999) 939-943. [28] J. Siepmann, N.A. Peppas, Modeling of drug release from delivery systems based on hydroxypropyl methylcellulose (HPMC), Adv. Drug Deliv. Rev., 48 (2001) 139-157.
23
[29] T.N. Hiew, D.L.H. Tan, Y.L. Tiang, P.W.S. Heng, Understanding the release performance of pellets with hydrophobic inclusions in sustained-release coating, Int. J. Pharm., 557 (2019) 229-237. [30] F. Theeuwes, Elementary osmotic pump, J. Pharm. Sci., 64 (1975) 1987-1991.
24
Table 1. Coating conditions and process parameters used to prepare the sustained release coated pellets. Coating conditions
For drug loading
For sustained release coating
Batch size (g)
1800
1800
Distance, nozzle from pellet bed (mm)
9
9
Spray nozzle protrusion (mm)
1
1
Spray nozzle diameter (mm)
0.8
1.2
Inlet air temperature (°C)
60-65
40
Atomizing air pressure (bar)
2.5
2.5
Spray rate (g/min)
6-10
6-10
25
Table 2. f2 values of pellets placed at different positions and cut to different cut-depths. Pellet position
f2 value
Pellet condition
f2 value
Bottom
39.4
Bisected
14.8
Lower quadrant
39.9
Cut in 0.2 mm depth pit
17.5
Middle
48.2
Cut in 0.3 mm depth pit
34.5
Upper quadrant
51.8
Cut in 0.4 mm depth pit
23.8
Top
38.6
Mid-radial
38.8
Top-radial
27.7
26
Figure 1: Schematic representation of preparation of SPIM.
27
Figure 2. Disintegration time of minitablets prepared with MCC (), MCC with 0.5%, w/w SSG () and MCC with 1%, w/w SSG ( ) across five compression forces.
28
(a)
(b)
Figure 3. (a) Drug release profiles of pellets compressed as SPIMs compared to uncompressed pellets (×). Pellet positions: Bottom (); lower quadrant (); middle (); upper quadrant (); top (); mid-radial; (); top-radial (). (b) T50% determined from the dissolution profiles of pellets shown in their relative locations on the cross-section of the SPIM.
29
(a)
(b)
Figure 4. (a) Schematic representation of the preparation of pellets with different cut depths. (b) Drug release profiles of the coated pellets (×) that were bisected (), cut in 0.2 mm deep pit (), 0.3 deep mm pit () and 0.4 mm deep pit ().
30
Figure 5. Schematic representation of the various release mechanisms involved in intact (I), punctured (II) and large cracked (III) coats, respectively.
31
Graphical abstract
32
S0939-6411(19)31305-0 https://doi.org/10.1016/j.ejpb.2019.11.006 EJPB 13182
To appear in:
European Journal of Pharmaceutics and Biopharmaceutics
Received Date: Revised Date: Accepted Date:
26 August 2019 17 November 2019 25 November 2019
Please cite this article as: T.N. Hiew, Y.H. Tian, H.M. Tan, P.W.S. Heng, A Mechanistic Understanding of Compression Damage to the Dissolubility of Coated Pellets in Tablets, European Journal of Pharmaceutics and Biopharmaceutics (2019), doi: https://doi.org/10.1016/j.ejpb.2019.11.006
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.
A Mechanistic Understanding of Compression Damage to the Dissolubility of Coated Pellets in Tablets
Tze Ning Hiewa, Yu Harn Tiana, Huei Ming Tanb, Paul Wan Sia Henga,*
aGEA-NUS
Pharmaceutical Processing Research Laboratory, Department of Pharmacy, National University of Singapore, 18 Science Drive 4, Singapore 117543, Singapore.
bEngineering
Science Programme, Faculty of Engineering, National University of Singapore, 9 Engineering Drive 1, Singapore 117575, Singapore.
*Corresponding author: P.W. S. Heng. Tel.: +65 65162930; fax: +65 67752265. Email: [email protected]
1
Abstract Damage to the drug diffusion coat barrier of controlled release pellets by the compaction force when preparing multiple-unit pellet system tablets is a major concern. Previous studies have shown that pellets located at the tablet axial and radial peripheral surfaces were more susceptible to damage when compacted due to the considerable shear encountered at these locations. Hence, this study was designed to assess with precision the impact of pellet spatial position in the compact on the extent of coat damage by the compaction force via a single pellet in minitablet (SPIM) system. Microcrystalline cellulose (MCC) pellet cores were consecutively coated with a drug layer followed by a sustained-release layer. Chlorpheniramine maleate was the model drug used. Using a compaction simulator, the coated pellets were compacted singly into 3 mm diameter SPIMs with MCC as the filler. SPIMs with individual pellets placed in seven positions were prepared. The uncompacted and compacted coated pellets, as SPIMs, were subjected to drug release testing. The dissolution results showed that pellets placed at the topradial position were the most susceptible to coat damage by the compaction force, while pellets positioned within the minitablet at the middle and upper quadrant positions showed the least damage. The SPIM system was found to be effective at defining the extent of coat damage to the pellet spatial position in the compact. This study confirmed that coated pellets located at the periphery were more susceptible to damage by compaction, with pellets located at the top-radial position showing the greatest extent of coat damage. However, if the pellet was completely encrusted by the cushioning filler, coat damage could be mitigated. Further investigations were directed at how the extent of coat damage impacted drug release. Interestingly, small punctures were found to be most detrimental to drug release whilst coats with large surface cuts did not completely fail. A damaged pellet coat has some self-sealing ability and failure is not total. Thus, this study provides a deeper understanding of the consequence of coat damage to drug release when sustained release coated pellets are breached.
Keywords: coated pellet, minitablet, microcrystalline cellulose, sustained release, compaction, coat damage
2
1.
Introduction
Controlled release dosage forms are very popular with patients as these dosage forms offer the advantages of reduced pill burden and dosing frequency, which could potentially manage patient noncompliance more effectively. Regulating the rate of drug release also impacts therapeutic efficacy as a gradual drug release in the gastrointestinal tract over a prolonged period provides better pharmacological outcomes. Multi-particulate dosage forms, commonly comprising of coated and uncoated pellets are advantageous as after administration, the pellets distribute uniformly in the gastrointestinal tract without high local concentration and are not prone to catastrophic dose dumping should some coated pellets fail. The key to developing multi-particulate dosage forms is the inclusion of slow release coated pellets. Pellet-based dosage forms are often delivered in gelatin capsules which can be contentious due to the use of animal protein. The solution is therefore the introduction of compacted pellets as multi-unit pellet system (MUPS) tablets. However, a common concern associated with MUPS tableting is the preservation of the integrity of the functional coat often used to modify drug release [1]. The film coatings used to modify drug release are usually acrylic- or cellulosic-based [2], with polymethacrylate and ethyl cellulose as the most commonly used polymers, respectively. The mechanical properties of the polymeric film are important for determining its durability and the ability to withstand the applied pressure during tableting. Due to their flexibility, acrylic polymers were reported to form more deformable films on pellets meant for compaction into MUPS tablets, whereas ethyl cellulose was more brittle by comparison [3].
In addition to the choice of coating polymer used, excipients as fillers are very important in MUPS tableting. Ideally, the filler should be able to cushion the coated pellets from compression force related damage and yet, make mechanically strong tablets which can readily disintegrate upon ingestion. The challenges presented in designing MUPS tablets require substantial research work at understanding how cushioning fillers can mitigate compaction-induced pellet coat damage and the conditions which may potentially aggravate or ameliorate the extent of damage. In studies using lactose, the particle size of the lactose was a critical attribute and micronized lactose was reported to be best [4, 5]. Highly
3
compressible excipients were also found to be good cushioning fillers [6-9]. Besides blending pellets with cushioning excipients, attempts were also made to directly deposit the cushioning excipients as an overcoat on the coated pellets [10-13].
The volumetric ratio between the filler and pellets or, critical pellet volume fraction [14] was reported to influence the extent of pellet coat damage. Critical pellet volume fraction values ranging from 0.374 [14] to 0.39 [4] were optimal to maintain a percolating network of cushioning fillers around the pellets, thereby minimizing pellet-pellet contacts during compression. In addition, the ratio between cushioning filler and pellets was also found to influence mechanical strength, friability and disintegration time of the resultant tablets [15].
While the coating polymer, cushioning filler and pellet volume fraction are critical attributes to successful MUPS tablet design, one area in MUPS tableting which can potentially impact release performance of the coated pellets but not well investigated or understood is the effect of pellet spatial position on coat damage. A previous study excavated pellets from different locations of the MUPS tablets and it was reported that the spatial location of a pellet within the tablet played an important role at determining the extent of compaction-related pellet coat damage [14]. Pellets located at the peripheral surfaces, be they axial or radial surfaces, had pellet coats more deformed or damaged, especially when compression pressure used was above the crushing strength of the pellets.
Herein, the effect of pellet spatial position in a compact on coat damage due to compression was investigated using a single pellet in minitablet (SPIM) system. With the SPIM system, pellets, singly or a few together, can be accurately placed in the desired position of the compact and spatial-related factors such as potential damage due to pellet-pellet or pellet-wall contacts and pellet-filler interactions could be better understood. In this study, pellets were placed singly at seven different positions and compacted into minitablets with microcrystalline cellulose as the cushioning filler. For SPIM pellets in contact with the tool surfaces, they were on top or bottom in contact with the punches and mid-radial or top-
4
radial in contact with the die wall. For pellets located centrally inside the tablet, they were placed at either the upper quadrant, middle or lower quadrant. Dissolution studies were conducted on the compacted pellets to determine the extent of pellet coat damage.
5
2. 2.1.
Materials and methods Materials
Chlorpheniramine maleate (CPM; BP grade, China) was used as the model drug in this study. Hydroxypropyl methylcellulose (HPMC; Methocel VLV, Dow Chemical, USA) was used as supplied for drug layering. Aqueous poly(meth)acrylates dispersion (ERS; Eudragit RS30D, Evonik, Germany) was used as the sustained-release coating polymer, with triethyl citrate (TEC; Merck, Germany) as the plasticizer. Talc (Chemipure, Singapore) was milled in a pin mill (ZM1000, Retsch, Germany) and used as an anti-tack agent in the sustained-release film coating. Microcrystalline cellulose (MCC) cores (500710 μm; Celphere CP-507, Asahi Kasei, Japan) were the starter cores for drug layering, followed by sustained release coating. Degassed purified water was the dissolution medium. MCC (Ceolus PH-101, Asahi Kasei, Japan) was the filler cushioning agent in the SPIM with sodium starch glycolate (SSG; DFE Pharma, Germany) as the disintegrant. Potassium dihydrogen phosphate (KH2PO4; Merck, Germany), ortho-phosphoric acid (H3PO4; Merck, Germany) and methanol (HPLC grade, Fisher Chemical, Canada) were used to prepare the mobile phase for high performance liquid chromatography (HPLC).
2.2.
Pellet coating
The MCC cores were consecutively coated with a drug layer followed by the sustained release layer in an air suspension coater (FlexStream module, MP-1, GEA, UK). For drug loading, 1.8 kg of MCC cores were layered with a coating media containing 7%, w/w CPM and 14%, w/w VLV HPMC to a final coat weight gain of 21%, w/w, calculated based on the percentage ratio of dry weight of coat material deposited onto the weight of the cores. The pellets were then oven dried overnight at 60 °C. The pellets were then fractionated using 500 and 1000 μm aperture size sieves (Endecotts, UK) to remove the fines and agglomerates, respectively.
The drug loaded 500-1000 μm pellet fraction was overcoated with poly(meth)acrylate polymer. The coating media consisted of 39.2%, w/w ERS, 2.4%, w/w TEC and 5.9%, w/w talc, which corresponded
6
to a final dry coat composition of 58.9%, w/w polymer, 11.7%, w/w TEC and 29.4%, w/w talc. The pellets were coated to 17.5%, w/w coat weight gain. Upon completion, the coated pellets were cured in-process for 30 min with water, and further cured for 12 h at 40 °C in a hot air oven. The pellets were then again fractionated to 500-1000 μm. The conditions and process parameters used for the coating of both the drug and sustained-release layers, optimized prior to this study, are shown in Table 1.
2.3. 2.3.1.
Physical characterization of pellets Pellet size and shape determination
Randomly selected pellets, about 200, were imaged (DP71, Olympus, Japan) using an optical microscope (SZ61, Olympus, Japan) and the captured images of 200 randomly selected pellets captured were analyzed (Image-Pro 6.3, Media Cybernatics, USA) for pellet mean Feret’s diameter, defined as the average caliper length of the pellet. Additionally, two shape determinant parameters, roundness and aspect ratio, were also calculated using Eqs. (1) and (2), respectively. Roundness reflects the sphericity of the pellets, while aspect ratio represents the extent of pellet elongation [16]. Therefore, the roundness and aspect ratio of a perfectly spherical pellet would be unity. 𝑝2 𝑅𝑜𝑢𝑛𝑑𝑛𝑒𝑠𝑠 = 4 × 𝜋 ×𝐴
𝐴𝑠𝑝𝑒𝑐𝑡 𝑅𝑎𝑡𝑖𝑜 =
(1)
𝑙 𝑏
(2)
where p is the perimeter, A is the area of the pellet, l and b are the major and minor axes of the ellipse equivalent to the particle, respectively.
2.3.2.
Mechanical testing of pellets
The crushing force of 20 pellets was determined by diametrical compression between two platens on a tensile tester (EZ Test, Shimadzu, Japan) fitted with a 100 N load cell and averaged. The crushing rate used was 2 mm/min, and the maximum load (Fm) required to crush a pellet was recorded. The tensile (crushing) strength of each pellet was calculated using Eq. (3):
7
𝐶𝑟𝑢𝑠ℎ𝑖𝑛𝑔 𝑠𝑡𝑟𝑒𝑛𝑔𝑡ℎ =
𝐹m 2
𝜋𝑑
(3)
×4
where d is the mean Feret’s diameter of the pellets, determined via optical microscopy.
2.4. 2.4.1.
Preparation of SPIM Optimization of compression pressure and filler composition
Powder blends of MCC and 0.5 or 1%, w/w SSG were compacted into 3 mm minitablets with a compaction simulator (STYL’One, Medelpharm, France) fitted with multiple-tip, standard concave tooling (8 tips per punch; Natoli Engineering Company, US). Control minitablets of only MCC were also prepared. A fill depth of 6 mm (1.5 × tablet diameter), which produced approximately 12 mg minitablets, was used. The minitablets were compacted to five pressures, per punch tip, were 34, 68, 102, 136 and 170 MPa. Disintegration tests (The United States Pharmacopeia method; DT2, Sotax, Switzerland) were carried out at 37 ± 2 °C in purified water with at least six randomly chosen minitablets and the results were averaged.
2.4.2.
Compression of SPIM on compaction simulator
2.4.2.1. Design of pellet placement guide for SPIM preparation A pellet placement guide (Figure 1) was designed to accurately insert a pellet centrally within the die cavity. Using an aluminum (6061 alloy) disc of 22.5 mm diameter and 7 mm thick, eight protrusions were milled, each with 1.0 mm through holes drilled. The design and fabrication work was accomplished with the use of 3D computer-aided design software (SOLIDWORKS 2018 SP02, Dassault Systèmes, France), computer-aided manufacturing software (HSMWorks Ultimate 2019, Autodesk, US), 3D desktop Computer Numerical Control machine (Nomad 883 Pro, Carbide 3D, US), 1/8” flat, ¼” corner rounding, 30° and 60° carbide end mills, and a 1.0 mm HSS drill bit.
8
2.4.2.2. Spatial pellet placements in SPIMs The coated pellets were compacted singly at 65 MPa using the equal force mode. Each SPIM contained approximately 12 mg of MCC pre-blended with 1%, w/w SSG as the disintegrant. In each SPIM, a pellet was placed at one of the seven pre-selected positions. For tablet surface placement, pellet was deposited at either the top, bottom, mid-radial or top-radial positions. For pellet centrally positioned inside the SPIM, it was placed using the placement guide at the lower quadrant, middle or upper quadrant of the tablet, assisted by bi-layer filling with pre-loading of 25%, 50% or 75%, respectively, of the total feed weight of the intended tablet feed, followed by the pellet and then the balance of the tablet weight (Figure 1).
Preparation of SPIM with pellet at the bottom was carried out by carefully placing a pellet centrally on the concavity of the power punch, followed by 12 mg of the pre-weighed blended filler. Conversely, SPIM with pellet at the top was prepared by first filling all the filler followed by careful placement of pellet centrally at the top with the aid of a placement guide.
For SPIMs containing pellets at the mid-radial and top-radial positions, an additional tamping step was added prior to pellet placement to ensure that the pellets could be placed firmly at the circumferential side of the die cavity, in contact with the die wall.
2.5. 2.5.1.
Drug release testing Optimization of conditions for drug release testing
A coloring agent (Eurolake Sunset Yellow, AJ4929, Netherlands) was hand-blended with MCC at a 1:1 gravimetric ratio for 2 min. The powder blend was then compacted into 100 mg compacts of 6 mm diameter using a single-station press (NP-RD10, Natoli Engineering Company, US) installed with standard concave tooling (Natoli Engineering Company, US). These compacts were then placed individually into 1.4 cm diameter test tubes containing 8 mL of deionized water maintained at 37 ± 0.5 °C by a thermostatically controlled shaker water bath. The shaker water bath was oscillated at 40, 60 or
9
80 oscillations per min, with regular visual inspection for homogeneity of the color distribution in the test tubes. After 15 min, 2 mL of the medium was withdrawn and replaced with an equal volume of deionized water, and the color distribution monitored for another 15 min. Based on visual inspection, 40 oscillations per min was found to be ideal and used for the subsequent drug release experiments.
2.5.2.
Drug release from individual pellets
The uncompacted and compacted coated pellets, as SPIMs, were subjected to drug release testing in deionized water. Each dosage form was placed in a test tube of 1.4 cm diameter containing 8 mL of deionized water maintained at 37 ± 0.5 °C in a shaker water bath pendulating at 40 oscillations per min. At time points of 15, 30, 60 and 120 min, 2 mL aliquots were withdrawn using a mechanical pipette and replaced with fresh media. After 120 min, the dosage form was crushed with a glass rod and the remnants left to shake for another 15 min to allow for complete dissolution of the drug, after which 2 mL of dissolution medium was withdrawn. Dissolution tests were performed with at least three randomly chosen pellets/tablets and the results were averaged. The drug content in each sample was determined using HPLC.
From the plot of cumulative drug release against time, T50%, the time required for 50% drug release, was determined. Additionally, similarity factor f2, which was calculated based on Eq.(4), was used to determine if the release profiles from uncompacted pellets and compacted pellets were similar. f2 is a logarithmic reciprocal square root transformation of the sum of squared error and is a measurement of the similarity in the percent (%) dissolution between the two curves. A f2 value between 50 and 100 indicates equivalence [17]. 1 𝑓2 = 50 × log {[1 + ( ) 𝑛
2.5.3.
𝑛
∑
|𝑅𝑖 ― 𝑇𝑖|2 ] ―0.5 × 100}
(4)
𝑖=1
High performance liquid chromatography
Drug contents were determined using a HPLC system (LC-2010C HT, Shimadzu, Japan) with a mobile phase of 50%, v/v phosphate buffer (H3PO4/KH2PO4, pH 3.0, 50 mM) and 50%, v/v methanol delivered, 10
isocratic, at 1 mL/min through a 100 mm × 4.6 mm C18 column (ACE Generix, 5 μm, Advanced Chromatography Technologies, UK) maintained at 40 °C. A wavelength of 260 nm was employed for detecting CPM at its retention time of 3.25 min.
2.6.
Preparation of pellets with different cut depth
Deep transverse cuts were made to randomly selected pellets to simulate compaction damage on pellet coatings. Reproducible linear cuts were enabled by accurately drilled seat holes of various depths on an aluminum plate. The fabricated plate is a 100 mm × 50 mm × 3 mm thick 6061 aluminum alloy plate, drilled with 1.0 mm diameter holes, of various depths from 0.2-1.0 mm, in 0.1 mm increments in depths, and hole pitch of 10 mm. Design and fabrication was carried out as per pellet placement guide (Section 2.4.2.1). The drilling process was done using a micro solid end mill (905L010-MEGA-T, SECO Tools, Sweden), programmed with 10,000 rpm spindle speed and 20 mm/s plunge feed rate.
Each test pellet was inserted in a hole of target cut depth and vertically bisected from the top with the use of a sharp surgical blade until the blade edge was flush with the surface of the aluminum plate. Selected pellets were also completely bisected to mimic a completely broken pellet. At least five pellets were prepared for each cut depth, and drug release rates determined.
11
3. 3.1.
Results Physical characterization of pellets
The Feret’s diameter, roundness and aspect ratio of the coated pellets were 735.0 ± 42.3 μm, 1.14 ± 0.08 and 1.10 ± 0.07, respectively, from pellet images and by image analysis. Using diametrical compression, the crushing strength of the pellets was found to be 33.6 ± 4.3 MPa.
3.2.
Disintegration time of minitablets
The disintegration time of the SPIM was expected to influence the rate of drug release, as nondisintegrating SPIMs could potentially confound drug dissolution from the coated pellets. SPIMs containing pellets may be encrusted by the filler and are susceptible to lower or no drug release despite being in contact with the dissolution medium. Non-disintegrating minitablets is especially a concern with MCC as it was previously reported to exert a shielding effect that slowed down drug release [18].
As shown in Figure 2, disintegration time of the minitablets increased with compression pressure. Without disintegrant aid, they failed to disintegrate even after 30 min when compacted at 170 MPa. An addition of SSG enabled all the minitablets to disintegrate within 30 s. The contribution of SSG to reduce disintegration time was especially marked at high compression pressures of 136 MPa and 170 MPa. For minitablets compacted at 136 MPa, disintegration time shortened to 20.4 s and 12.6 s when 0.5%, w/w and 1%, w/w SSG were added, respectively. Hence, MCC with 1%, w/w SSG was chosen for the formulation of SPIM.
3.3.
Effect of pellet position on drug release
Dissolution studies on the uncompacted and compacted pellets, as SPIM, were carried out to elicit the impact of pellet position in the compact on its rate of drug release. As the pellets were coated with a sustained-release polymer, any compromise to coat integrity would be expected to result in faster drug release. Drug release profiles of the pellets over 120 min, both uncompacted and compacted, are shown in Figure 3a.
12
Regardless of the position of the pellet in the SPIM, all the compacted pellets showed faster drug release compared to the uncompacted pellets, reflecting some damage to the sustained release coat caused by the compaction force. This was further substantiated by the f2 values (Table 2a), which showed that with the exception of pellets placed at the upper quadrant position, the drug release profiles of the other compacted pellets were significantly different from that of the uncompacted pellets. However, differences in the rate of drug release were observed for the different SPIMs, suggesting that pellets placed at different positions were damaged to different extents. Pellets placed at the middle and upper quadrant positions showed the least damage. There was increased severity of damage but to rather similar extents for pellets placed at the top, bottom, lower quadrant and mid-radial positions. Pellets placed at the top-radial position was most damaged, as reflected by fastest drug release rate and dissolution T50% (Figure 3b) averaging 37.3 min, or about 2.5 times shorter than that of the uncompacted pellets (92.8 min).
3.4.
Effect of pellet cut depth on drug release
Observation from the dissolution studies of the coated pellets placed at different positions within the SPIM appeared to show different degrees of coat damage but unexpectedly, not a total failure. As shown in Figure 3a, drug release profiles can broadly be sub-divided into 3 clusters. In order to better understand the findings of this study, coated pellets with different cut depths were prepared to mimic the severity of coat damage by the extent of crack formation caused by compression during tablet making. By depositing the coated pellets in pre-drilled 1 mm diameter pits of depths, 0.2 mm, 0.3 mm and 0.4 mm, circumferential slits were made on the pellets using a sharp razor, bisecting the exposed portion. By computation based on the mean Feret’s diameter of the pellets and the depth of the hole, the cuts could be translated to approximately 65, 56 and 47% cracks along the circumference of the pellets placed in 0.2 mm, 0.3 mm and 0.4 mm pits, respectively.
Figure 4b shows the dissolution profiles of the cut pellets. Pellets that were bisected showed the fastest drug release, with a T50% of 16.9 min, and 90% of the drug was released within an hour. This was 13
followed by deepest cut coated pellets in 0.2 mm depth pit, which had a T50% of 18.8 min, close to that of the fully bisected pellets. Pellets cut in 0.3 mm deep pit had a longer T50% of 47 min. Interestingly, pellets cut in the 0.4 mm deep pit, with the smallest fraction of coat damage, had a rather different drug release profile, showing a sigmoidal pattern. The amount of drug released from these pellets during the first 15 min was comparable to that of the uncompacted pellets, suggesting that the puncture did not caused any immediate change in release rate but a latent burst release occurred which pointed to a rather catastrophic consequence of coat failure, as the later high release rate paralleled that of a fully bisected coated pellet.
14
4.
Discussion
Fabrication of tablets invariably involves the application of compression pressure to a powder bed to consolidate the particles placed within a confined space into a coherent mass of fixed geometry [19]. When the pressure is applied, the particles first undergo rearrangement which results in further volume reduction and an increase in compact porosity. When the interparticulate frictional forces are too high for further volume reduction, the particles will undergo deformation [19]. Studies have shown that the distribution of compact density and mechanical properties across the matrix of a tablet are unequal. In flat face tablets, despite equal force being applied over the powder surface by the upper and lower punches, variability had been reported [20, 21]. Therefore, it is not unexpected that the forces experienced during fabrication and resultant density distribution in a biconvex tablet is nonhomogeneous. Density distribution analyses of biconvex tablets showed that the tablet density was in general the lowest around the center of the tablet, and gradually increased laterally towards the edge of the tablet, where density was the highest [22-26]. This was attributed to the localized densification of the powder as the biconvex shape resulted in thinner volume along the peripheral edge. This nonhomogeneity in density distribution will undoubtedly occur within a MUPS tablet and potentially be translated to locality-dependent differences in the forces exerted on the pellets during compaction. Indeed, pellets located at both the radial and axial surfaces within MUPS tablets were found to suffer a larger extent of deformation and coat damage due to direct impact and friction with the compression tools [14]. Building on the reported observations, this study employed the use of SPIM to more precisely examine the extent of pellet coat damage in relation to the spatial position of the pellets and the ultimate consequence to the damaged pellets in their performance as sustained drug release vehicles.
The dissolution results from the SPIMs showed that pellet spatial position played a definitive role in the extent of coat damage. Pellets placed at the top-radial position was most damaged, as evident from the shortest dissolution T50% as compared to pellets in other locations. As biconvex minitablets were made, the pellets at the top-radial position received the highest impact forces from compression pressure exerted along with friction against the compaction tools, which collectively inflict considerable damage to the sustained release coat. The other pellets in contact to tablet surfaces, at the top, bottom and mid15
radial positions, were also damaged but slightly less than pellets at the top-radial position. Nonetheless, the drug release profiles suggest rather severe damage to the functional coat, impairing somewhat their diffusion barrier function. Interestingly, pellets located at the lower quadrant of the SPIM and encased by the filler material had rather similar release profiles as those located at the top, bottom and midradial positions. In contrast, pellets located at the middle and upper quadrant positions were least damaged and had similar drug release profiles. As the SPIMs were compacted by simultaneously moving the upper and lower punches in opposing direction, the forces exerted biaxially should theoretically be close. The consequence of this mechanical operation was two-fold. Firstly, the compression pressure exerted by the lower punch was more sustained compared to the upper punch, as although the latter moved rapidly downwards, it was immediately withdrawn once the set thickness or maximum compression pressure was achieved. With the recorded dwell time ranging from 1-5 ms, the maximum compression pressure event was very transient. Secondly, examination of the upper and lower punch pressures revealed a disparity between them. From the compaction data of the SPIMs, the pressures exerted by the lower punch were consistently higher, by 8-16% higher than that of the upper punch. This observation would explain the higher degree of coat damage encountered by the pellets in the lower quadrant of the SPIMs as compared to those in the middle or upper quadrant.
The compaction force used to make the SPIMs in this study was 3.8 kN. With a cup area of 7.35 mm2, the compression pressure was 65 MPa, which was approximately twice the average crushing force of the pellets when they were crushed between two platens. This compression pressure was chosen to be equivalent to a 10 kN force applied on the powder bed with 14 mm flat faced tooling. Even at a rather high compression pressure applied, it was surprising to discover that the compacted coated pellets did not fail completely, and appearing to show gradated degrees of failure. It was originally assumed that a burst rapid release would result from a fractured pellet coat, and emulating the release profile of a bisected coated pellet, with more than 90% drug released within an hour. However, as a gradated degree of failure was observed for the compacted pellets, it could be hypothesized that some re-sealing mechanism existed, possibly by the hydrated polymeric gel layer underneath the sustained release coat. A previously published study examined the effect of pellet coat rupture on drug release by piercing the 16
pellet coat with a needle. Microscopic observation revealed a shift in the pellet core to an eccentric position brought about by the pressure build-up within the pellet, especially in the region away from the puncture point. The movement of the core material consisting of a hydrated polymeric gel and insoluble MCC core sealed off the puncture point, thereby trapping the drug within the pellet [27]. In addition, the swelling HPMC polymer also aided in re-sealing the punctured coat, thereby serving as an additional diffusion barrier for the drug.
While pellets with a small puncture exhibited self-sealing and preserving abilities, it was not known if this phenomenon would prevail with respect to compaction created cracks, often linear. The outcome from the dissolution studies showed that even with the cut coats, the pellets exhibited some cut depthdependent delay in drug release. Despite deep cuts made on the coats of the pellets, complete failure of the dosage form akin to a dose dump was not seen. However, the f2 values suggested that the drug release profiles were still significantly different from the uncompacted pellets (Table 2b). This finding shows some innate ability of the HPMC polymer to re-seal the breached sustained release coat. HPMC is known to swell when wetted, particularly with higher viscosity grades used in sustained release matrix tablets. HPMC may also erode and dissolve [28]. Therefore, the ability of HPMC to control and sustain drug release requires a fine balance between its rate of gel formation, swelling and erosion. In this study, the drug was dissolved in the VLV HPMC solution before it was coated onto the MCC cores. As the HPMC used for drug loading was of a very low viscosity grade, it is postulated that any gel formed could not possess sufficient gel strength to effectively seal the crack in the coat. Regardless, since the rate of drug release showed some correlation to the extent of coat damage, HPMC certainly had played a role in limiting drug release during dissolution.
Another hypothesis to explain the results is that in breached sustained release coat, the two breached edges could realign themselves and continue to act as a continuous diffusion barrier. For a typical coated pellet, drug release could proceed by the drug concentration gradient built-up across the barrier, from the internal drug reservoir to the bulk medium [29]. Thus, a tear to the diffusion coat may not be as
17
detrimental to its barrier properties, provided no material leakage or differential pressure build up to disrupt the status quo.
Eudragit RS30D contains water-insoluble copolymers of ethyl acrylate, methyl methacrylate and a low content of a methacrylic acid ester with quaternary ammonium groups (trimethylammonioethyl methacrylate chloride). The ammonium groups are present as salts and make the polymer coat permeable through pore formation when wetted. This allows water permeation into the pellet to dissolve and leach out the drug by diffusion (Figure 5-I). The acrylic diffusional barrier enables the formation of a saturated drug reservoir of dissolved CPM within the pellet, with constant drug diffusion across the water-insoluble, semi-permeable polymer coat. The reservoir of CPM within the pellet maintains a constant concentration gradient across the diffusional barrier. In the case of cut coats, faster drug release was observed due to some content leakage (Figure 5-III).
Interestingly, pellets with the shallower cuts exhibited a greater degree of burst release. This is counterintuitive as the pellets with the least damage should theoretically leverage most on the re-sealing ability of the breached coats to retard drug release. It is postulated that HPMC swelled when wetted and had contributed to the immediate re-sealing of the slit. However, with time, the osmotic pressure buildup occurred due to the presence of HPMC in the coated pellets. With an intact coat, there is isotropic osmotic pressure build-up, thus resulting in uniform swelling. However, with partially damaged pellet coat, there is anisotropic osmotic pressure build-up, with higher pressure accumulated away from the crack, which results in uneven swelling [27]. The internal pressure would push the core material content towards the breached opening at the cut of the coat. This effectively produced a weak osmotic pump system, propelling some pellet content out via the cut to equilibrate the osmotic pressure build-up within and release character is akin that of an osmotic pump system (Figure 5-II). With a larger crack, any osmotic pressure build-up would be less [30]. Thus, for the smallest cut, pellets cut in the 0.4 mm depth pit, had initial retarded drug release as shown by the lag phase but eventually, the osmotic pressure build-up was the release driver, which provided a steady rapid ejection of pellet liquid content along
18
with the drug. This drastic latent increase in drug release produced a profile much identical to that of the bisected pellets but the early lag phase was rather similar to that of the intact coated pellet.
By comparing the release profiles of the pellets from SPIMs to the pellets with different cut depths, it was observed the pellets placed at the top-radial position experienced a greater extent of coat damage compared to the other pellets, with the extent of damage similar to pellets cut in the 0.2 mm deep pit. The release profiles of the other pellets from SPIMs were more similar to the pellets cut in the 0.3 mm deep pit. The difference in drug release from pellets with different cut depths helped to further unravel the consequences of compaction-induced coat damage on the drug release profiles obtained from the compacted coated pellets as SPIMs. Clearly, pellet coat damage by compaction does not necessarily cause complete product failure as there could be differences in release rates as attributed to the different extents of breaches or cracks present. Even for pellets completely surrounded by the cushioning filler, their spatial position and extent of damage inflicted could be determinant of their sustained drug release function but may be in a rather complex manner.
19
5.
Conclusion
In this study, the SPIM system was devised and found beneficial for evaluating the effect of pellet spatial position on the extent of coat damage. Pellets located at the periphery were more susceptible to damage by compaction, with pellets located at the top-radial position showing the highest extent of coat damage. However, if the pellet was surrounded completely by the cushioning filler, the degree of coat damage could be mitigated. Additionally, the extent of pellet coat damage was also found to affect the re-sealing capability of the damaged coat. Therefore, even with the use of compression pressure above the crushing force of the pellets, the pellets did not fail completely in their sustained drug release function. In MUPS tablets, the pellets are randomly distributed within the tablet matrix and the ensemble drug release profile obtained represents a complex cacophony of release profiles by individual pellets located in the MUPS tablet. The role played by the pellet spatial position on pellet coat damage emphasizes the importance of formulating MUPS tablets below the critical pellet volume fraction, as this would reduce the number of pellets located at the surface of the tablets, thereby ensuring that the damage caused to the pellet coat could be kept to a minimum. The findings in this study provide important insights into the consequences of coat damage in MUPS tableting and clearly, a breached diffusion coat may have some degree of self-sealing properties.
20
Acknowledgements This research work was supported by the GEA-NUS PPRL fund (N-148-000-008-001). The authors also wish to acknowledge support from Asahi Kasei, Japan for the gratis supply of the Ceolus PH-101 and Celphere CP-507. The initial concept of SPIM arose from the project work supported by Asahi Kasei in MUPS tableting (R-148-000-275-597).
21
References [1] S.R. Béchard, J.C. Leroux, Coated pelletized dosage form: Effect of compaction on drug release, Drug Dev. Ind. Pharm., 18 (1992) 1927-1944. [2] L.A. Felton, P.B. O'Donnell, J.W. McGinity, Mechanical properties of polymeric films prepared from aqueous dispersions, in: J.W. McGinity, L.A. Felton (Eds.) Aqueous Polymeric Coatings for Pharmaceutical Dosage Forms, CRC Press, Boca Raton, FL, 2008, pp. 105-128. [3] R. Bodmeier, O. Paeratakul, Mechanical properties of dry and wet cellulosic and acrylic films prepared from aqueous colloidal polymer dispersions used in the coating of solid dosage forms, Pharm. Res., 11 (1994) 882-888. [4] W.C. Chin, L.W. Chan, P.W.S. Heng, A mechanistic investigation on the utilization of lactose as a protective agent for multi-unit pellet systems, Pharm. Dev. Technol., 21 (2016) 222-230. [5] X. Lin, W.C. Chin, K.-f. Ruan, Y. Feng, P.W.S. Heng, Development of potential novel cushioning agents for the compaction of coated multi-particulates by co-processing micronized lactose with polymers, Eur. J. Pharm. Biopharm., 79 (2011) 406-415. [6] G.J. Vergote, F. Kiekens, C. Vervaet, J.P. Remon, Wax beads as cushioning agents during the compression of coated diltiazem pellets, Eur. J. Pharm. Sci., 17 (2002) 145-151. [7] F. Zeeshan, K.K. Peh, Y.T.F. Tan, Exploring the potential of a highly compressible microcrystalline cellulose as novel tabletting excipient in the compaction of extended-release coated pellets containing an extremely water-soluble model drug, AAPS PharmSciTech, 10 (2009) 850. [8] J.J. Torrado, L.L. Augsburger, Effect of different excipients on the tableting of coated particles, Int. J. Pharm., 106 (1994) 149-155. [9] S.U. Kucera, J.C. DiNunzio, N. Kaneko, J.W. McGinity, Evaluation of Ceolus™ microcrystalline cellulose grades for the direct compression of enteric-coated pellets, Drug Dev. Ind. Pharm., 38 (2012) 341-350. [10] X. Pan, M. Chen, K. Han, X. Peng, X. Wen, B. Chen, J. Wang, G. Li, C. Wu, Novel compaction techniques with pellet-containing granules, Eur. J. Pharm. Biopharm., 75 (2010) 436-442. [11] S.A. Altaf, S.W. Hoag, J.W. Ayres, Bead compacts. I. Effect of compression on maintenance of polymer coat integrity in multilayered bead formulations, Drug Dev. Ind. Pharm., 24 (1998) 737-746. [12] S.A. Altaf, S.W. Hoag, J.W. Ayres, Bead compacts. II. Evaluation of rapidly disintegrating nonsegregating compressed bead formulations, Drug Dev. Ind. Pharm., 25 (1999) 635-642. [13] A. Hosseini, M. Körber, R. Bodmeier, Direct compression of cushion-layered ethyl cellulosecoated extended release pellets into rapidly disintegrating tablets without changes in the release profile, Int. J. Pharm., 457 (2013) 503-509.
22
[14] M. Xu, P.W.S. Heng, C.V. Liew, Formulation and process strategies to minimize coat damage for compaction of coated pellets in a rotary tablet press: A mechanistic view, Int. J. Pharm., 499 (2016) 2937. [15] A. Debunne, C. Vervaet, D. Mangelings, J.-P. Remon, Compaction of enteric-coated pellets: influence of formulation and process parameters on tablet properties and in vivo evaluation, Eur. J. Pharm. Sci., 22 (2004) 305-314. [16] A.M. Bouwman, J.C. Bosma, P. Vonk, J.A. Wesselingh, H.W. Frijlink, Which shape factor(s) best describe granules?, Powder Technol., 146 (2004) 66-72. [17] Food and Drug Administration, Guidance for Industry: Dissolution Testing of Immediate Release Solid Oral Dosage Forms, Center for Drug Evaluation and Research (CDER), (1997). [18] T.N. Hiew, M.I.B. Alaudin, S.M. Chua, P.W.S. Heng, A study of the impact of excipient shielding on initial drug release using UV imaging, Int. J. Pharm., 553 (2018) 229-237. [19] G. Alderborn, Tablets and compaction, in: M.E. Aulton, K.M.G. Taylor (Eds.) Aulton's Pharmaceutics: The Design and Manufacture of Medicines, Elsevier, Edinburgh, 2013, pp. 504-549. [20] A. Kandeil, M.C. de Malherbe, S. Critchley, M. Dokainish, The use of hardness in the study of compaction behaviour and die loading, Powder Technol., 17 (1977) 253-257. [21] D. Sixsmith, D. McCluskey, The effect of punch tip geometry on powder movement during the tableting process, J. Pharm. Pharmacol., 33 (1981) 79-81. [22] I.C. Sinka, J.C. Cunningham, A. Zavaliangos, Analysis of tablet compaction. II. Finite element analysis of density distributions in convex tablets, J. Pharm. Sci., 93 (2004) 2040-2053. [23] Y. Hattori, R. Aoki, M. Otsuka, Use of partial least-squares analysis and fractionated X-ray computed tomography images in the investigation of density distribution of round tablets, Powder Technol., 302 (2016) 261-264. [24] R.K. May, K. Su, L. Han, S. Zhong, J.A. Elliott, L.F. Gladden, M. Evans, Y. Shen, J.A. Zeitler, Hardness and density distributions of pharmaceutical tablets measured by terahertz pulsed imaging, J. Pharm. Sci., 102 (2013) 2179-2186. [25] A. Djemai, I.C. Sinka, NMR imaging of density distributions in tablets, Int. J. Pharm., 319 (2006) 55-62. [26] B. Eiliazadeh, K. Pitt, B. Briscoe, Effects of punch geometry on powder movement during pharmaceutical tabletting processes, Int. J. Solids Struct., 41 (2004) 5967-5977. [27] P.W.S. Heng, L.W. Chan, S.H. Chew, Mechanism of pellet coat rupture and its effect on drug release, Chem. Pharm. Bull. (Tokyo), 47 (1999) 939-943. [28] J. Siepmann, N.A. Peppas, Modeling of drug release from delivery systems based on hydroxypropyl methylcellulose (HPMC), Adv. Drug Deliv. Rev., 48 (2001) 139-157.
23
[29] T.N. Hiew, D.L.H. Tan, Y.L. Tiang, P.W.S. Heng, Understanding the release performance of pellets with hydrophobic inclusions in sustained-release coating, Int. J. Pharm., 557 (2019) 229-237. [30] F. Theeuwes, Elementary osmotic pump, J. Pharm. Sci., 64 (1975) 1987-1991.
24
Table 1. Coating conditions and process parameters used to prepare the sustained release coated pellets. Coating conditions
For drug loading
For sustained release coating
Batch size (g)
1800
1800
Distance, nozzle from pellet bed (mm)
9
9
Spray nozzle protrusion (mm)
1
1
Spray nozzle diameter (mm)
0.8
1.2
Inlet air temperature (°C)
60-65
40
Atomizing air pressure (bar)
2.5
2.5
Spray rate (g/min)
6-10
6-10
25
Table 2. f2 values of pellets placed at different positions and cut to different cut-depths. Pellet position
f2 value
Pellet condition
f2 value
Bottom
39.4
Bisected
14.8
Lower quadrant
39.9
Cut in 0.2 mm depth pit
17.5
Middle
48.2
Cut in 0.3 mm depth pit
34.5
Upper quadrant
51.8
Cut in 0.4 mm depth pit
23.8
Top
38.6
Mid-radial
38.8
Top-radial
27.7
26
Figure 1: Schematic representation of preparation of SPIM.
27
Figure 2. Disintegration time of minitablets prepared with MCC (), MCC with 0.5%, w/w SSG () and MCC with 1%, w/w SSG ( ) across five compression forces.
28
(a)
(b)
Figure 3. (a) Drug release profiles of pellets compressed as SPIMs compared to uncompressed pellets (×). Pellet positions: Bottom (); lower quadrant (); middle (); upper quadrant (); top (); mid-radial; (); top-radial (). (b) T50% determined from the dissolution profiles of pellets shown in their relative locations on the cross-section of the SPIM.
29
(a)
(b)
Figure 4. (a) Schematic representation of the preparation of pellets with different cut depths. (b) Drug release profiles of the coated pellets (×) that were bisected (), cut in 0.2 mm deep pit (), 0.3 deep mm pit () and 0.4 mm deep pit ().
30
Figure 5. Schematic representation of the various release mechanisms involved in intact (I), punctured (II) and large cracked (III) coats, respectively.
31
Graphical abstract
32