Noboru Hoshi

When drugs are encapsulated, electrification (the electrostatic charge of the capsule) may sometimes cause problems, such as capsule adhesion during transportation or dispersion of the capsule content in the filling process. In this article, the electrification of gelatin and nongelatin capsules is examined. 

Polyvinyl alcohol (PVA) copolymer capsules are a form of nongelatin capsule under development. PVA, acrylic acid (AA) and methyl methacrylate (MMA) are used as raw materials. As previously reported, these capsules have advantages, such as low gas permeability, and can be particularly suitable for encapsulation of hydrophilic solvents, such as polyethylene glycol (PEG) 400, and surfactants.

  Using such capsules facilitates the formulation of insoluble drugs and is expected to enhance bioavailability. 

The electrostatic characteristics of PVA copolymer capsules were determined using gelatin and nongelatin (hypromellose (HPMC)) capsules as controls. After, the stability of a PVA copolymer capsule preparation of indomethacin (IND) solution was examined as a model of an insoluble drug preparation. In addition, the bioavailability of a sparingly water-soluble drug, acetaminophen, and a practically water-insoluble drug, nifedipin (NF), in both gelatin and nongelatin capsules was evaluated by administering the preparations to Beagle dogs. 

It is well known that electrification of capsules has different characteristics depending on the raw material of the polymer compound used. Low electrification is desirable and there have been many cases where an antistatic agent was used to reduce capsule electrification. The electrostatic characteristics of three types of capsules were determined. 

Measurement of electrification using the Faraday cage method. 

Approximately 2–3 g of each capsule type was measured out and placed in a Faraday cage (Advantest; Tokyo, Japan). The amount of triboelectricity generated by the capsules was measured with a digital multimeter (Advantest) and the specific charge electrification (nC/g) was calculated (Table 1). Of the three capsule types, the gelatin and HPMC capsules showed high electrification values with positive electricity. In contrast, the PVA copolymer capsule had the lowest electrification value with negative electricity. 

Table 1: Specific charges for the three capsule types per 1 g. 

Measurement of surface potential charges after generation of triboelectricity. 

Approximately 10 g of each capsule type was measured out, placed in a tablet friability tester. (Tsutsui Rikagaku Kikai; Tokyo, Japan). Triboelectricity was generated for 1 min and the surface potential charges of 20 randomly selected capsules were then measured using a surface charge electrometer (Model 1279, Monroe Electronics; NY, NY, USA) 5 and 10 min after the start of the test (Figure 1). 

Figure 1: Surface potential charge for the three capsule types. 

The gelatin capsules gained the most charge and were characterized by positive electrification. Additionally, their surface potential did not decrease easily. HPMC capsules are known to have less water content (~3%) than gelatin capsules (~15%) and their electrification tends to be lower. This tendency was confirmed in these experiments. However, the HPMC capsules showed a high level of surface potential, even after 10 min, and the charge tended not to attenuate easily. This phenomenon is considered to be a result of the low moisture content of the capsules. 

The PVA copolymer capsules displayed the lowest level of electrification, as well as a negative charge. After displaying a surface potential charge of ~ –500 V immediately after electrification, the capsules attenuated toward 0 V, proving that these capsules are characterized by easy attenuation of the surface potential charge.  

The attenuation of the surface potential charge of the capsules was affected by the functional groups of the raw materials and their polymeric structure. The PVA copolymer capsules are not easily electrified and show easy attenuation of any electricity that is generated. 

In contrast with gelatin and HPMC capsules, it has been proven that PVA copolymer capsules are compatible with PEG 400, Tween 80 and LABRASOL.

  In a further experiment, the stability of a liquid preparation of insoluble PEG 400 encapsulated in PVA copolymer capsules was evaluated. 

 A PVA capsule of size 0 was filled with 625 mg of PEG 400 solution containing 25 mg of IND and band-sealed. 

 The encapsulated sample was placed in an airtight Lamizip AL (Seisannipponsha Ltd; Japan), a heat-shielded PET/AL/PE multilayer film bag, and an accelerated stability test was performed at 40 °C and 75% relative humidity. No appearance changes, leaks or softening were observed after 1, 2 and 3 months of storage. The encapsulated solution was collected from the capsule and its water content was measured according to the Karl Fischer's method. It increased only slightly on storage, but this did not influence the drug dissolution rate (Figure 2). 

Figure 2: Change in water content of the encapsulated solution. 

The dissolution testing of the capsule preparation was conducted using Japanese Pharmacopoeia (JP) Apparatus 2 (Toyama Sangyo Co., Ltd; Osaka, Japan [paddles]) at 50 rpm in water and two test solutions (first and second fluids for the dissolution test of JP). IND concentration was assayed using ultraviolet (UV) absorbance at 320 nm. Figure 3 shows that IND dissolved quickly in water and in the second fluid. The dissolution rate for all of the capsule preparations at 15 min was ~100%. However, in the first fluid only 90% of the drug was released. This is considered to be a result of decreased solubility of IND in acid. No effect of storage on drug dissolution was found in any of the fluids. 

Figure 3: IND release rate from a solution filled into a PVA copolymer capsule. 

Administration of an encapsulated powder form of a water-soluble drug

 Size 1 gelatin, HPMC and PVA copolymer capsules were filled with 250 mg of a mixture containing 150 mg of lactose powder mixed with 100 mg of AA. 

 The drug dissolution testing of each capsule was evaluated according to the 

Guideline for Bioequivalence Studies of Generic Products

 using the JP Apparatus 2 (paddles) at 50 rpm in water, 25% McIlvaine buffer solution (pH 5.5, 6.5) and two test solutions (the first and second fluids for the dissolution test of JP). The AA concentration in the fluids for the dissolution test of JP was assayed using UV absorbance at 245 nm. 

Figure 4: AA release rate from a capsule preparation in various test solutions. 

AA encapsulated in the gelatin, HPMC and PVA copolymer capsules dissolved quickly in the dissolution test solutions. In addition, as the difference in the dissolution rates after 15 min was within ±10% among all the capsules, they were deemed to be equivalent (Figure 4 and Table 2). 

Table 2: AA release rate* from a capsule preparation in various test solutions (%). 

 Animal studies were performed using Beagle dogs (6 dogs; all male; aged 1–2 years; weight 10–11 kg; the same individual used in all tests) with a drug-free period of <1 week between tests. 

The capsules were administered orally under fasting and nonfasting conditions, and AA concentration in plasma was assayed by HPLC — as reported previously — before and after 0.25, 0.5, 1.0, 1.5, 2, 3, 4 and 6 h after administration.

  The AA plasma concentration profile is shown in Figure 5. 

Figure 5: AA blood concentration curve upon administration of the three capsule types filled with 100 mg AA to Beagle dogs (n=56). 

 (time to the maximum drug concentration) were measured from AA plasma concentration data and the area under the curve (AUC) was calculated according to the trapezoid method. The difference among the preparations on each parameter was evaluated by the paired t-test (Table 3). There was no significant difference (p>0.05) between the capsules in each pharmacokinetic (PK) parameter, and the gelatin and nongelatin capsules filled with AA, which was used as a model of a water-soluble drug, were deemed to be biologically equivalent. 

Table 3: PK parameters upon administration of the three capsule types filled with 100 mg AA to Beagle dogs (n=56). 

Administration of an encapsulated insoluble drug formulation

 100 mg of a 150 mg mixture containing 140 mg of lactose mixed with 10 mg of NF was placed in a size 1 gelatin capsule and used for the dissolution test. In animal studies, a capsule filled with 150 mg of mixture containing 120 mg of lactose mixed with 30 mg of NF was used to ensure adequate absorption. In addition, 0.5 mL of a PEG 400 solution containing 10 mg NF was encapsulated in a PVA copolymer capsule to form a band-sealed liquid formulation. 

 Drug dissolubility from the three capsule preparations was evaluated using JP Apparatus 2 (paddles) at 50 rpm in water and the first fluid for the dissolution test of JP. The NF concentration in the fluid for the dissolution test of JP was assayed using UV absorbance at 238 nm. 

The dissolution profile is shown in Figure 6. Dissolution of NF from the PVA copolymer capsule filled with NF/PEG solution in the first fluid and in water was ~100% after 15 min, similar to the drug dissolution rate without a capsule. Conversely, the dissolution of NF from the gelatin capsule filled with powder was significantly delayed (<40% in 45 min). 

Figure 6: Drug release from preparations containing NF 10 mg. 

 A solution containing 10 mg of NF, a PVA copolymer capsule preparation of the solution containing of 10 mg of NF, and a gelatin capsule preparation of the powder containing 30 mg of NF were administered orally to fasting Beagle dogs. The NF concentration in plasma, as reported previously, was determined before and after 0.5, 1.0, 2.0, 3.0, 4.0, 6.0 and 8.0 h after administration and was assayed using HPLC-ECD.

  The solution was administered directly to the stomach 

The NF plasma concentration profile is shown in Figure 7 and the PK parameters are shown in Table 4. There were no significant differences in PK parameters between the encapsulated and non-encapsulated solutions (p>0.05). In addition, when the solution preparation and the powder preparation were compared taking the difference in NF doses into consideration, the former was found to be improved approximately 6-fold in terms of C

 and approximately 3-fold regarding AUC. 

In terms of bioavailability, PVA copolymer and HPMC capsules should be treated as equivalent to the widely used gelatin capsules (Figure 5 and Table 3). In addition, PVA copolymer capsules enable encapsulation of hydrophilic solvent, which makes insoluble drugs dissolve, formulations (Figure 7 and Table 4). These solution preparations were confirmed to enhance bioavailability when administered to Beagle dogs. 

Figure 7: NF blood concentration curve after administration of the NF preparations to Beagle  dogs (n=56). 

PVA copolymer, which is a new pharmaceutical agent, completed safety studies based on the relevant guidelines and was registered in the FDA Drug Master File in 2006. 

Table 4: PK parameters upon administration of the NF preparations to Beagle dogs (n=56). 

These PVA copolymer capsules have various advantages compared with traditional gelatin capsules, such as low water content, low electrification level and the capability of filling with hydrophilic solvent, and are expected to facilitate the development of new drug formulations in the future. 

 was Manager of Technological Research at the Research & Development Group, Nisshin Kasei Co., Ltd (Japan).

 is a reseacher at the Research & Development Group, Nisshin Kasei Co., Ltd (Japan).

 is Head of Oral Solid Formulation at CMC Development Laboratories, Shionogi & Co., Ltd (Japan).

 is Scientist, Oral Solid Formulation at CMC Development Laboratories, Shionogi & Co., Ltd (Japan).

Articles

When drugs are encapsulated, electrification (the electrostatic charge of the capsule) may sometimes cause problems, such as capsule adhesion during transportation or dispersion of the capsule content in the filling process.

Creating PVA copolymer capsules

Hard capsules have been developed as an edible container to mask the taste and odour of medicines. Traditionally used for powder or granulated formulations, capsules have also been adapted to contain oily liquids, tablets and even powders for inhalation. They are popular because of their relative ease of manufacture (compared with other dosage forms such as tablets) and their flexibility to accommodate a range of fill weights. Additionally, capsules readily demonstrate bioequivalence between different strengths of the same formulation.	

The solubility of many compounds used in potential new drugs is very low because they are selected for their affinity to receptors, which increases as the lipophilicity of a compound increases. Although these compounds are expected to have a high clinical performance, they often fail to become new drug entities because of their low absorption in the gastro-intestinal (GI) tract - a result of poor dissolution. 	

The authors developed polyvinyl alcohol (PVA) acrylic acid methyl methacrylate polymer (PVA copolymer) capsules following research indicating that the bioavailability of some insoluble drugs was enhanced when dissolved in macrogol 400 (polyethyleneglycol [PEG] 400), but conventional hard capsules could not tolerate macrogol 400. An earlier study reported that the solubility of a drug in macrogol 400 is approximately 100 times higher than in water (Table I).1 When the drug/water mixture was administered in a suspension to dogs, only 2.2% of the dose was absorbed, but when administered in a solution of macrogol 400, absorption was 91%.

Table I:  Solubility of insoluble drugs in macrogol 400 and PK parameters after administration in dogs.	

This suggests that pharmaceutical manufacturers could develop dosage forms of insoluble drugs with macrogol 400, improving the solubility of such entities. Because the formulations and manufacturing processes are simple, no scale-up studies would be required - possibly reducing drug development times.	

Because of the large potential of capsules that can hold macrogol 400, the authors developed new capsules, synthesizing materials that are suitable as capsule shells. By copolymerizing acrylic acid (AA) and methyl methacrylate (MMA) on PVA as a skeleton and then using the obtained PVA copolymer as capsule shells, the authors successfully developed capsules that can be filled with macrogol 400.

 In this paper, the physical properties of the PVA copolymer, and the characteristics and pharmaceutical applications of PVA copolymer capsules, are investigated. 	

Initially, the authors examined why conventional capsule shells do not tolerate macrogol 400. When gelatin capsules were filled with macrogol 400, they became brittle and broke easily because the moisture in the shell was absorbed by the macrogol. When hydroxypropyl methylcellulose (HPMC) capsules were filled with macrogol 400, the agent oozed out through the capsule shell. 				

The authors believed that new synthetic polymers would be more suitable for capsule shell materials rather than natural polymers or polysaccharides. Thus, different polymers were synthesized, using styrene resin, polyurethane, acrylic polymer and chitosan as a skeleton. 	

PVA copolymer was developed by the authors (Figure 1). Commercially available, partially-saponified PVA was used as the raw material for the skeleton of the copolymer. AA and MMA were emulsion-polymerized onto the PVA skeleton.	

Figure 1:  Typical synthesis of PVA copolymer.				

Dissolution of PVA copolymer. Capsules made only of PVA are available, although they are easily softened by surrounding moisture. In the PVA copolymer, MMA was used to increase the hardness of the capsule shell; however, increasing the amount of MMA decreases the polymer solubility. Thus, AA was copolymerized to increase the solubility at neutral pH. The composition ratios of PVA, AA and MMA in the PVA copolymer can be modified; the best copolymer is formed when the levels of PVA, AA and MMA are 70–80%, 2.5–5.0% and 15–25%, respectively.	

Drug capsules should dissolve in purified water, as well as in simulated gastric fluid (pH 1.2) and simulated intestinal fluid (pH 6.8) of the disintegration test method listed in the Japanese Pharmacopoeia (JP). The dissolution of PVA copolymer cast film in the above media was examined. The result showed that the film was soluble in all three fluids, indicating that the copolymer has suitable dissolution characteristics. The film showed no erosion, swelling or dissolution in macrogol 400.	

Film strength. PVA copolymer film was formed in 100 μm thickness using the casting method. The breaking strength and elongation rate of 40310 mm film segments were examined (Table II), which showed the breaking strength of the PVA copolymer film to be 32.2 N/mm

. The value was slightly lower than that of gelatin film (55.8 N/mm

), but comparable to that of HPMC film (30.8 N/mm

); therefore, the PVA copolymer film was considered acceptable for practical use. 

Table II:  Tensile strength of films and Table III:  Gas permeability through films.	

When the PVA copolymer film was moistened in a chamber (25 Â°C/75% relative humidity [RH]), its strength decreased by approximately 25%. Gelatin and HPMC specimens showed larger reductions (more than 50%) in strength when moistened at the same condition. 	

Although the gelatin and HPMC films showed a low elongation rate before and after moistening, the PVA copolymer film showed a markedly higher rate before and particularly after moistening (312% of the original rate).	

Gas permeability of the film. The 100 μm thick film was also used to examine gas permeability by the American Society for Testing and Materials (ASTM) method. Water vapour permeability through PVA copolymer film at 25 Â°C/90% RH was 323.2 g/m2/24 h, which was between the values of the gelatin and HPMC films (Table III). There was no marked difference in water vapour permeability between the three films. In contrast, oxygen permeability through PVA copolymer was significantly less than through gelatin and HPMC films, indicating that it should be impermeable to the PVA copolymer film.	

Moisture absorption and desorption isotherm. Generally, water-soluble polymers start to absorb moisture when relative moisture increases by more than 70% and the absorption rate dramatically increases when relative moisture exceeds 80%. The moisture absorption isotherm curve of PVA copolymer is similar to that of gelatin (Figure 2). The moisture desorption of PVA copolymer is similar to its absorption isotherm curve, whereas, for gelatin film, the desorption rate is slower than its absorption rate. 	

Figure 2:  Isotherms of moisture absorption and desorption at 25ÃÂ°C.	

Safety study. The authors developed hard capsules using the copolymer based on the properties described earlier. Because the PVA copolymer is a new additive, it is necessary to examine its safety. Recently, the number of newly developed additives has decreased because testing their safety is too expensive. In this case, Huntingdon Life Sciences (UK) performed the safety testing. A single administration test in rats; a micronucleus study in mice; an Ames test; a human lymphocyte chromosome aberration study; a single administration study in dogs; and a maximum tolerated dose (MTD) toxicity study were completed, all indicating no safety concerns. The single administration study in rats showed that the oral lethal dose was more than 2000 mg/kg. 	

No significant changes in autopsy, pathological or pharmacological findings were observed when the copolymer was administered continuously for 14 days at a dose of 500 mg/kg in dogs. Studies of the absorption, distribution, metabolism and excretion of 

C-labelled PVA copolymer are currently under way. After these are completed, a preliminary long-term oral administration study and a subacute study of the repeated administration for 3–6 months will be performed in rats. 	

PVA copolymer capsules were prepared by the dipping and forming method. Carrageenan (0.05-0.5%) was added as a gelling agent and potassium chloride (0.05-0.5%) was added as a gelling promoter. This method requires no additional investment for capsule manufacturers because conventional gelatin capsule manufacturing machines can be used. Prototype PVA copolymer capsules were coloured, showed a good gloss and were not different from conventional capsules. 				

Disintegration and dissolution. Prototype PVA copolymer capsules (size #0) were filled with macrogol 400 and the time taken for the contents to begin to leak out was measured (Table IV). The capsules were filled with the disintegrant croscarmellose sodium and the paddle method (50 rpm) was used to measure the disintegration time as an index of the start of dissolution. The capsules opened in less than 10 min in all the media. 		

Table IV:  Disintegration test and lag time in dissolution test of PVA copolymer capsules.	

Capsule hardness . The relationships between the brittleness and moisture content of PVA copolymer, gelatin and HPMC capsules were compared using a hardness tester (Shionogi Qualicaps, Japan). A 50 g weight was dropped at a height of 10 cm onto a capsule and the percentage of broken capsule was determined (Figure 3). At a water content of 8%, the gelatin capsules became brittle and the percentage of broken capsules was 100%. The HPMC capsules did not break, even at 1% water content. PVA copolymer capsules became brittle when water content was less than 4%, but were less brittle compared with gelatin capsules. PVA copolymer capsules did not become brittle or break easily, even when filled with macrogol 400 (which absorbs moisture from the capsule shell) or silica gel grains (desiccant). The authors concluded that these test conditions were too severe to estimate capsule brittleness.		

To evaluate capsule deformation, the load required to deform each capsule by 50% was measured using a load cell (Figure 4). The results illustrate that PVA copolymer capsules can be deformed by moisture absorption, but this can be prevented by controlling the water content of the filling and the humidity of the environment, or with moisture-proof packaging. 		

Figure 3:  Brittleness of capsules at different water content and Figure 4:  Load required to deform 50% of capsules at different water content.	

Filling dissolution. When dissolution testing PVA copolymer capsules (size #2) filled with macrogol 400, the bottom of the capsules quickly dissolved (2 min) in purified water, JP 1st fluid (pH 1.2) and JP 2nd fluid (pH 6.8), with the macrogol 400 leaking out from the bottom. 	

Two sets of capsules for the insoluble drugs tolbutamide and indomethacin were prepared. One capsule was filled with the drug as a solution of macrogol 400 and the other was filled with a mixture of the drug, lactose (as a filler) and croscarmellose sodium (as a disintegrant). The dissolution behaviour of the capsules was compared using the JP paddle method (50 rpm). The tolbutamide/macrogol 400 capsules almost dissolved completely in 10 min in all the test solutions. However, for the capsules filled with the drug, filler and disintegrant, only 80% of the drug had dissolved after 60 min (Figure 5). 			

Figure 5:  Improvement of tolbutamide capsule (size #2) dissolution when filled with macrogol 400 and Figure 6:  Improvement of indomethacin capsule (size #2) dissolution when filled with macrogol 400.	

Indomethacin in macrogol 400 solution quickly dissolved in all of the three test fluids, as observed with tolbutamide (Figure 6). However, the dissolution rate of indomethacin with filler and disintegrant was slow, particularly in JP 1st fluid where the solubility was low and the dissolved percentage at 60 min was approximately 20%. 	

Based on the above data, the dissolution of encapsulated insoluble drugs can be improved when they are dissolved in a solution of macrogol 400.	

Absorption of indomethacin in rats. Indomethacin PVA copolymer capsules were prepared using either a solution of macrogol 400 or a mixed powder formulated with lactose and croscarmellose sodium. Both sets were filled into mini-capsules (size #9) and administered to rats to compare the plasma profiles of the drug (Figure 7).

 The results illustrate that between 180-360 min, the capsule containing the drug as a solution of macrogol 400 demonstrated a higher plasma concentration (8 μg/mL) compared with the capsules containing the mixed powder (1 μg/mL). The data suggest that PVA copolymer capsules filled with macrogol 400 improve the bioavailability of insoluble drugs.  		

Figure 7:  Indomethacin absorption improvement after administration of macrogol solution to rats and Figure 8:  Stability of PVA copolymer capsules.	

Tolerance of PVA copolymer capsules. The solubilizing agents macrogol 400, Tween 80 and Labrasol were filled into PVA copolymer capsules and stored in accelerating conditions (40 Â°C/75% RH) in a sealed container to examine the tolerance of the capsules to the agents. The capsules showed no change in appearance after 6 months (Figure 8).	

When a solubilizing agent with a high water content is filled into a capsule, the moisture causes the capsule shell to soften and/or deform. This can be prevented by controlling the water content of the agent and the humidity of the manufacturing environment, and by using moisture-proof packaging, as done for conventional capsules. 	

Comparison with conventional capsules. Current commercially available hard capsules are made of gelatin or HPMC. Table V lists the advantages and disadvantages of PVA copolymer capsules compared with these capsules. PVA copolymer capsules have similar advantages to HPMC capsules because both have been developed to overcome the drawbacks of gelatin capsules. The advantages of PVA copolymer capsules include no animal-derived material, a low water content, no Maillard reaction and a low electrostatic propensity. Additionally, PVA copolymer capsules demonstrate the unique properties of having very low oxygen permeability and the ability to contain macrogol 400. 	

A recent report suggests that the successful development of one or two new drugs requires approximately 100 discovery projects and initial in vitro screening of approximately 7 million compounds.

 After in vitro screening, pharmacological and preclinical safety studies are performed followed by clinical studies to confirm the efficacy and safety of the drugs in humans before being marketed. 				

The development of high throughput screening and information technology are making the selection of candidate compounds at the initial screening stage extremely efficient. However, subsequent research still requires trial and error experiments; when compounds are extremely insoluble or unstable, a large number of such experiments are necessary that may lengthen the drug development time and increase the costs.	

Figure 9 shows how PVA copolymer capsules can help develop insoluble drugs. With capsule drug development, the formulation should remain constant between preclinical and Phase IV studies. This avoids issues with changing bioequivalence because of formulation changes and scale-up. Additionally, it may be easier to develop a formulation having a high bioavailability. PVA copolymer capsules, therefore, are favourable for the quick and low cost development of insoluble drugs.	

Figure 9:  Efficient development of insoluble drugs with solution-filled capsules.	

PVA copolymer does not have chemically active groups such as those in gelatin and it is more tolerable to different agents than gelatin and HPMC capsules. Before PVA copolymer can be used in marketed drugs, further safety studies are required and the authors anticipate that this will take a further 2 years.				

Patents of the capsule have already been filed and the authors plan to supply PVA copolymer capsules globally.

 The authors also intend to develop equipment that will give the capsules a protective oxygen barrier.	

It is said that more than 200 billion capsules are currently used globally. More than 100 years have passed since gelatin capsules were first developed. HPMC capsules were developed at the end of the 20th century,

and the authors hope that PVA copolymer capsules will help drug development in the 21st century.		

Table V:  Comparison of capsule characteristics.	

The bioavailability of some insoluble drugs is enhanced when they are dissolved in the solubilizing agent macrogol 400, although conventional hard capsules cannot tolerate the agent. This article investigates a PVA copolymer, which has been developed by the authors, examining its properties and its suitability as a material in capsule formulations.