Oral Delivery of Niclosamide as an Amorphous Solid Dispersion That Generates Amorphous Nanoparticles during Dissolution

Niclosamide is an FDA-approved anthelmintic that is being studied in clinical trials as a chemotherapeutic and broad-spectrum antiviral. Additionally, several other applications are currently in the preclinical stage. Unfortunately, niclosamide is a poorly water soluble molecule, with reduced oral bioavailability, which hinders its use for new indications. Moreover, niclosamide is a poor glass former; in other words, the molecule has a high tendency to recrystallize, and it is virtually impossible to generate a stable amorphous solid employing the neat molecule. Previously, our group reported the development of an amorphous solid dispersion (ASD) of niclosamide (niclosamide ASD) that generates nanoparticles during its dissolution, not only increasing niclosamide’s apparent solubility from 6.6 ± 0.4 to 481.7 ± 22.2 µg/mL in fasted state simulated intestinal fluid (FaSSIF) but also its oral bioavailability 2.6-fold in Sprague–Dawley rats after being administered as a suspension. Nevertheless, niclosamide ASD undergoes recrystallization in acidic media, and an enteric oral dosage form is needed for its translation into the clinic. In this work, we further characterized the nanoparticles that generated during the dissolution of the niclosamide ASD. Cryogenic transmission electron microscopy (Cryo-TEM) and wide-angle X-ray scattering (WAXS) revealed that the nanoparticles were amorphous and had a particle size of ~150 nm. The oral dosage forms of niclosamide ASD were formulated using commercial enteric capsules (Capsuline® and EudracapTM) and as enteric-coated tablets. The enteric dosage forms were tested using pH-shift dissolution and acid-uptake tests, using the USP type II dissolution apparatus and the disintegration apparatus, respectively. The capsules exhibited a higher percentage of weight gain, and visual rupture of the Capsuline capsules was observed. Eudracap capsules protected the formulation from the acidic media, but polymer gelling and the formation of a nondispersible plug were noted during dissolution testing. In contrast, enteric-coated tablets protected the formulation from acid ingress and maintained the performance of niclosamide ASD granules during dissolution in FaSSIF media. These enteric-coated tablets were administered to beagle dogs at a niclosamide dose of 75 mg/kg, resulting in plasma concentrations of niclosamide higher than those reported in the literature using solubilized niclosamide at a higher dose (i.e., 100 mg/kg). In summary, an enteric oral dosage form of niclosamide ASD was formulated without hindering the generation of nanoparticles while maintaining the increase in the niclosamide’s apparent solubility. The enteric-coated tablets successfully increased the niclosamide plasma levels in dogs when compared to a niclosamide solution prepared using organic solvents.

apeutic, antibacterial, and broad-spectrum antiviral (including SARS-CoV-2), among other indications, as described in several published review papers [24,25]. The niclosamide ASD formulation increased niclosamide's oral bioavailability in rats. However, niclosamide in the ASD undergoes recrystallization if exposed to acidic media [26]. Consequently, the oral administration of niclosamide ASD would be inefficient without being included in a proper enteric dosage form. This can be an applicable to other ASDs composed of weakly acidic APIs and/or polymers. As an example, using a weakly acidic polymer, Elkhabaz et al.
(2019) observed a drastic drop in the supersaturation when running pH-shift tests using an ASD composed of posaconazole (weakly basic API) and HPMC-AS (weakly acidic polymer). During the acid phase, the polymer could not protect the drug from recrystallization at higher drug loadings, leading to a poor dissolution performance of the ASD [27].
In this work, we aimed (i) to further understand the solid-state properties of the nanoparticles that were generated during dissolution; (ii) to select a proper enteric (delayed release) dosage form for the delivery of niclosamide ASD; and (iii) to conduct in vivo pharmacokinetic testing in beagle dogs. Our hypothesis was that an enteric-coated dosage form would protect the formulation from recrystallization without hindering the generation of the nanoparticles, therefore increasing the oral bioavailability of niclosamide.

Preparation of the Niclosamide ASD Granules
Niclosamide anhydrate was purchased from Shenzhen Nexconn Pharmatechs LTD (Shenzhen, China); Kollidon ® VA64 (PVP-VA) and Kolliphor ® TPGS (D-α-tocopheryl polyethylene glycol succinate, TPGS) were obtained from BASF Corporation (Tarrytown, NY, USA). Niclosamide, PVP-VA, and TPGS were admixed at a 60:35:5 ratio and extruded using a ZSE 12 HP-PH (Leistritz Advanced Technologies Corp., Nuremberg, Germany). The extrusion was performed at a feed rate of 3 g/min with a screw speed of 50 rpm. The operating temperatures are specified in the Supplemental Information, Table S1. Then, granules of niclosamide ASD were obtained by milling the extrudates using a Fitzmill Comminutor Model L1A Benchtop Mill (The Fitzpatrick Company, Westwood, MA, USA) operating at 5000 rpm and equipped with a 0.06" round hole screen.

Preparation of the Enteric-Coated Tablets of Niclosamide ASD
The enteric-coated tablets were prepared by first admixing 66.63% niclosamide ASD granules, 13.1% microcrystalline cellulose (Ceolus ® PH-102, Asahi Kasei Chemical Corporation, Tokyo, Japan), 8.8% croscarmellose sodium (Kiccolate TM , Nichirin Chemical Industries, LTD., Dojima, Japan), 0.28% magnesium stearate (Spectrum Chemical, New Brunswick, NJ, USA), and 11.1% sodium chloride (Fisher Chemical, Ottawa, ON, Canada). The blend was filled into a hopper placed on a Stoke F4 Press. The tablet cores were produced using a 10 mm round standard cup tooling and with a target final tablet core weight of 428 mg. Then, the tablet cores were added to a Vector LDCS-Hi Coater (Vector Corporation, Marion, IA, USA). Opadry II Clear (Colorcon, West Point, PA, USA) dispersion in either water or 20:80 water/ethanol mixture was used for seal coating to a target weight gain of 3%. Then, the sub-coated tablets were enteric coated with either a completely water-based coating (Acryl-EZE ® 93 A, Colorcon, West Point, PA, USA) or a mixture of water and ethanol-based coating (Opadry ® Enteric, Colorcon, West Point, PA, USA) to a target 10% weight gain.

HPLC Analysis
The samples from the dissolution tests were measured using a Dionex Ultimate 3000 HPLC system (Thermo Fisher Scientific Inc., Sunnyvale, CA, USA) at a 1.5 mL/min flow rate, and the UV detector was set at 331 nm and employing a ZORBAX SB-C18 column (4.6 × 250 mm, 5 µm) (Agilent Technologies, Palo Alto, CA, USA). The mobile phase was composed of (A) 0.3% formic acid in water solution and (B) acetonitrile (Fisher Scientific, Pittsburgh, PA, USA) mixed in a 40:60 ratio.

Dissolution Testing of the Granules, Capsules, and Tablets
The niclosamide ASD granules, capsules, and tablets underwent dissolution tests using a USP type II apparatus (Hanson SR8-Plus apparatus, Hanson Research Co., Chatsworth, CA, USA) with low volume vessels (150 mL) at 37.0 ± 0.5 • C and a paddle speed of 100 rpm. The biorelevant medium FaSSIF (biorelevant.com Ltd., London, UK) was used to simulate the intestinal dissolution medium present in humans. It was prepared according to the instructions from the manufacturer, using a buffer pH 6.5, composed of sodium hydroxide, monobasic sodium phosphate (Fisher Scientific, Fair Lawn, NJ, USA), and sodium chloride (Sigma Aldrich, Saint Louis, MO, USA).
A 230 mg sample of niclosamide ASD granules (equivalent to 80 mg of niclosamide) underwent the dissolution test in 150 mL of FaSSIF. In the case of the oral dosage forms, enteric capsules were placed inside sinkers to maintain them on the bottom of the vessel. Tablets, on the other hand, did not require sinkers. The samples were collected and filtered using a 25 mm 0.2 µm Nylon syringe filter (VWR International LLC, Radnor, PA, USA) for analysis by HPLC. For the pH-shift dissolution test, the dosage forms were first added to 120 mL of 0.1 N HCl for 2 h. Then, the tablets/capsules were transferred to 150 mL of FaSSIF medium. Testing of the Eudracap capsules required the modification of the testing, increasing the FaSSIF buffer pH to 6.8 due to the increased pH required to dissolve the capsule based upon the shell polymer chemistry.

Acid Uptake Test
To determine the acid resistance of the enteric oral dosage forms, six capsules/tablets were weighed and placed into a QC-21 disintegration test system (Hanson Research, Chatsworth, CA, USA) containing 0.1 N HCl at 37 • C for 2 h. Sinkers were used in the case of capsules. After the test, the capsules/tablets were blotted to remove the surface liquid and weighed again to determine the weight increase.

Polarized Light Microscopy (PLM)
The niclosamide ASD granules were placed on a glass slide and observed using an Olympus BX-53 polarized light microscope (Olympus Corporation of the Americas, Center Valley, PA, USA) with a first-order red compensator and 10× and 20× objectives. The pictures were taken using a QImaging QICAM digital camera (QImaging, Surrey, BC, Canada).

Wide-Angle X-ray Scattering (WAXS)
The niclosamide anhydrate was prepared by heating the niclosamide raw material at 100 • C for 15 min [28]. This must be performed due to the hygroscopic nature of niclosamide. Niclosamide Ha (hydrate) was prepared by pouring niclosamide raw material in water and then letting it dry at ambient temperature for 24 h under a hood.
Wide-angle X-ray scattering (WAXS) was employed to determine the crystallinity of the nanoparticles generated during dissolution in FaSSIF. After 2 h of dissolution, 5 mL were collected and added into a Float-A-Lyzer ® Device (Repligen, Rancho Dominguez, CA, USA) for dialysis. The sample was dialyzed using 1 L of water supersaturated with niclosamide (added in excess) for 24 h. This was conducted to remove the excess inorganic salts and bile salts from the sample and to increase the measurement resolution. Then, the sample was collected and immediately frozen using liquid nitrogen. Thereafter, the sample was lyophilized using an SP VirTis Advantage Pro shelf lyophilizer (SP Industries, Inc., Warminster, PA, USA) to obtain a dry powder. The primary drying process was at −40 • C for 20 h, and then the temperature was linearly increased to 25 • C over 20 h, followed by holding the temperature at 25 • C for 20 h. The pressure was maintained at 100 mTorr during the lyophilization process. The samples were mounted at the sample position inside the equipment using a sandwich of Scotch ® Magic Greener Tape and Scotch ® Permanent Double-Sided Tape (3M, Saint Paul, MN, USA), similar to the methodology used by Hövelmann et al. (2019) [29]. The samples were analyzed using a SAXSLab instrument (WAXSLab, Northampton, MA, USA) equipped with a PILATUS3 R 300 K detector (DECTRIS Ltd., Philadelphia, PA, USA). The instrument was operated using the Ganesha instrument software (SAXSLab, Northampton, MA, USA).
In addition, WAXS was used to analyze the potential recrystallization of the encapsulated niclosamide. The samples of the enteric capsules underwent 2 h of dissolution testing in acid medium (0.1 N HCl); then, they were opened and the content was dried for 24 h at ambient temperature under a hood. In the case of the Capsuline sample, the dried material was milled to obtain a powder. The sodium bicarbonate and Explotab ® were analyzed as received. The powder samples were added to 2 mm glass capillaries (Hampton Research, Aliso Viejo, CA, USA). Each sample was analyzed using a 2 mm off-centered beam stop mask for an acquisition time of 600 s.

Cryogenic Transmission Electron Microscopy (Cryo-TEM)
The samples of niclosamide ASD were prepared using the same methodology of dialysis described in the WAXS section. These liquid samples were added into a 20 s glow discharged Q2/2R 200 mesh copper grid and vitrified in ethane using a Vitrobot Mark IV (FEI, Millsboro, OR, USA) at 100% relative humidity. The samples were observed using a Glacios Cryo-TEM with Falcon 4 (Thermo Scientific, Waltham, MS, USA). The magnification was 92,000X, and the dose rate was~2.02e − /Å 2 s. The images were visualized, and their contrast was increased using the ImageJ software, version 1.53t (NIH, Bethesda, MR, USA).

Administration of Niclosamide ASD to Beagle Dogs
In vivo testing in fasted male beagle dogs (8 kg, 7 to 10 months old), following oral administration of enteric-coated tablets containing niclosamide ASD at a dose of 75 mg/kg, was performed. The dog study was conducted under approval from the ITR Laboratories Animal Care and Use Committee. The OLAW animal welfare number is F16-00201 (A5834-0). The animals were kept in an environment with a temperature of 21 ± 3 • C, relative humidity of 50 ± 20%, and 12 h light and 12 h dark. Following each dose administration, the animals were administered 5 mL of water to facilitate swallowing. 1 mL of blood was collected from the animals, and the samples were processed into plasma by centrifugation at 2500 rpm for 10 min at approximately 4 • C. The samples were analyzed by HPLC/MS. The limit of quantification was 40 ng/mL.

Niclosamide ASD Generated Amorphous Nanoparticles in Biorelevant Media
As previously reported, the niclosamide ASD formulation generates nanoparticles during its dissolution in FaSSIF biorelevant media; these particles were observed even after filtering the dissolution test media through 0.2 µm filters [26]. However, initially, their crystallinity was not determined. In theory, these nanoparticles should result from a phenomenon called liquid-liquid phase separation (LLPS) or glass-liquid phase separation (GLPS), where a drug-rich phase is formed when the drug concentration exceeds its amorphous solubility [10,13].
Cryo-TEM was conducted to directly observe these drug-rich metastable regions generated by niclosamide ASD in FaSSIF, as shown in Figure 1, as other more common techniques, such as nanoparticle tracking analysis and scanning electron microscopy, were unable to successfully resolve the nanoparticles. Surprisingly, the nanoparticles were spherical, similar to the ones previously reported in the literature [11,12,30]. Furthermore, the particles were surrounded by a bilayer membrane structure of approximately 4.5 nm ( Figure 1C). These are in agreement with the bilayer structures generated by FaSSIF that were reported by Clulow et al. (2017). They proved that FaSSIF media forms vesicles of 30-50 nm with bilayer membranes of approximately 5 nm in thickness using cryo-TEM. The bilayer membrane had a similar thickness to that measured for the niclosamide ASD particles [31]. Previously, we reported FaSSIF particles of~66 nm, measured by dynamic light scattering (DLS) [26].  [32][33][34]. These varieties of nanostructures were identified as vesicles, spherical-mixed micelles, disk micelles, and liposomes by cryo-TEM [31,34].
We recently reported that the niclosamide ASD nanoparticles exhibited a mean size of ~100 nm, and cryo-TEM showed particles sizes of approximately ~136 nm [26]. In addition, Harmon et al. (2016) reported interesting behavior of ASDs prepared with increasing concentrations of TPGS. Specifically, increasing concentrations of TPGS facilitated the disintegration of the ASD and led to a greater number of nanoparticles with reduced particle size. TPGS was found on the surface of nanoparticles and acted as a stabilizer [11]. Likewise, in our system, the bile salts in the FaSSIF media plus the TPGS in the ASD could facilitate the formation and stabilization of these nanoparticles. The formation of the nanoparticles did not happen to the same extent without FaSSIF (Supplemental Information, Figure S1). This combination of TPGS and biorelevant media led to an increased limit of congruency and simultaneous release of niclosamide and PVP-VA, as described by Saboo et al. (2021). Interestingly, this effect on the LoC was also observed in prototype formulations using different polymers at different drug loadings of niclosamide, including with other surfactants (Supplemental Information, Figure S2).  We recently reported that the niclosamide ASD nanoparticles exhibited a mean size of 100 nm, and cryo-TEM showed particles sizes of approximately~136 nm [26]. In addition, Harmon et al. (2016) reported interesting behavior of ASDs prepared with increasing concentrations of TPGS. Specifically, increasing concentrations of TPGS facilitated the disintegration of the ASD and led to a greater number of nanoparticles with reduced particle size. TPGS was found on the surface of nanoparticles and acted as a stabilizer [11]. Likewise, in our system, the bile salts in the FaSSIF media plus the TPGS in the ASD could facilitate the formation and stabilization of these nanoparticles. The formation of the nanoparticles did not happen to the same extent without FaSSIF (Supplemental Information, Figure S1). This combination of TPGS and biorelevant media led to an increased limit of congruency and simultaneous release of niclosamide and PVP-VA, as described by Saboo et al. (2021). Interestingly, this effect on the LoC was also observed in prototype formulations using different polymers at different drug loadings of niclosamide, including with other surfactants (Supplemental Information, Figure S2).
After confirmation and elucidation of the nanoparticle morphology by cryo-TEM, further solid-state characterization of the nanoparticles was performed to determine if the particles were amorphous resulting from LLPS or if they were crystalline formed from supersaturation during dissolution. As seen in Figure 2, WAXS confirms that the nanoparticles were amorphous with only crystalline peaks from the FaSSIF matrix being detected in the sample. No crystalline peaks from the niclosamide anhydrate or niclosamide Ha (hydrate form A) were detected. Importantly, niclosamide is characterized as a poor glass former, which means that it has a high tendency for recrystallization [35]. Niclosamide tends to recrystallize by forming strong hydrophobic π-π interactions [36]. Consequently, Figures 1 and 2 demonstrate that niclosamide can avoid recrystallization by generating these amorphous nanoparticles that can act as reservoirs to increase its oral bioavailability. This formulation strategy can facilitate the development of amorphous solid dispersions containing other poor glass formers. However, these efforts cannot be achieved if the ASD is not effectively translated into a finished dosage form that can be administered to patients. As such, the next challenge is to prevent the recrystallization of niclosamide in the ASD when exposed to acidic conditions that occurs as the dosage form transits through the stomach.
After confirmation and elucidation of the nanoparticle morphology by cryo-TEM, further solid-state characterization of the nanoparticles was performed to determine if the particles were amorphous resulting from LLPS or if they were crystalline formed from supersaturation during dissolution. As seen in Figure 2, WAXS confirms that the nanoparticles were amorphous with only crystalline peaks from the FaSSIF matrix being detected in the sample. No crystalline peaks from the niclosamide anhydrate or niclosamide Ha (hydrate form A) were detected. Importantly, niclosamide is characterized as a poor glass former, which means that it has a high tendency for recrystallization [35]. Niclosamide tends to recrystallize by forming strong hydrophobic π-π interactions [36]. Consequently, Figures 1 and 2 demonstrate that niclosamide can avoid recrystallization by generating these amorphous nanoparticles that can act as reservoirs to increase its oral bioavailability. This formulation strategy can facilitate the development of amorphous solid dispersions containing other poor glass formers. However, these efforts cannot be achieved if the ASD is not effectively translated into a finished dosage form that can be administered to patients. As such, the next challenge is to prevent the recrystallization of niclosamide in the ASD when exposed to acidic conditions that occurs as the dosage form transits through the stomach.

Niclosamide ASD Dissolution Performance and Stability Were Impaired When Formulated as an Enteric Capsule Formulation
We previously reported that niclosamide ASD undergoes recrystallization when tested in the pH-shift dissolution test [26]. Figure 3 shows how the ASD granules quickly recrystallized after being wetted by 0.1 N HCl, with the first signs of crystallization taking place after just 1 min. It was concluded that an enteric dosage form was required for a viable pharmaceutical product. This reduction in niclosamide's solubility in acidic conditions was reported by Yu et al. (2021) when testing niclosamide-montmorillonite hybridbased formulations [37]. Niclosamide is a weak acid and has an ionizable phenolic -OH group, with reported pKa values ranging from 5.6 to 7.2 [38][39][40][41]. Because of this, the solubility of niclosamide increases at higher pH values and its LogD decreases [38,42]. Needham (2021) experimentally increased niclosamide's concentration from 2.53 µ M at pH 3.66

Niclosamide ASD Dissolution Performance and Stability Were Impaired When Formulated as an Enteric Capsule Formulation
We previously reported that niclosamide ASD undergoes recrystallization when tested in the pH-shift dissolution test [26]. Figure 3 shows how the ASD granules quickly recrystallized after being wetted by 0.1 N HCl, with the first signs of crystallization taking place after just 1 min. It was concluded that an enteric dosage form was required for a viable pharmaceutical product. This reduction in niclosamide's solubility in acidic conditions was reported by Yu et al. (2021) when testing niclosamide-montmorillonite hybrid-based formulations [37]. Niclosamide is a weak acid and has an ionizable phenolic -OH group, with reported pKa values ranging from 5.6 to 7.2 [38][39][40][41]. Because of this, the solubility of niclosamide increases at higher pH values and its LogD decreases [38,42]. Needham (2021) experimentally increased niclosamide's concentration from 2.53 µM at pH 3.66 to 703.5 µM at pH 9.63. In addition to these solubility and solid-state challenges, it was reported that niclosamide undergoes chemical degradation in acidic and in basic media due to the fact of hydrolysis [43][44][45]. It has also been reported that alkaline solutions favor niclosamide's photooxidation [40]. Niclosamide's degradation products are toxic to mononuclear cells and neuronal and alveolar cell lines and nontoxic to hepatic cell lines [46]. to 703.5 µ M at pH 9.63. In addition to these solubility and solid-state challenges, it was reported that niclosamide undergoes chemical degradation in acidic and in basic media due to the fact of hydrolysis [43][44][45]. It has also been reported that alkaline solutions favor niclosamide's photooxidation [40]. Niclosamide's degradation products are toxic to mononuclear cells and neuronal and alveolar cell lines and nontoxic to hepatic cell lines [46]. The niclosamide ASD granules were encapsulated in two different commercially available size 0 capsules: Capsuline ® (HPMCP/HPMC) and Eudracap TM enteric (HPMC/Eudragit ® polymers) capsules. These enteric capsules contain weakly acidic polymers that release their contents in the intestines, where the polymer is soluble and the capsule shell dissolves to release its payload [47]. In addition, it has been reported that PVP-VA gelling, which is contained in the ASD composition, hinders the disintegration of the dosage forms and their drug release. Gelling can be delayed by adding inorganic salts into the formulation [23]. Therefore, sodium bicarbonate was added for two reasons. First, the bicarbonate ion is a kosmotrope and delays the polymer gelling, and second, it increases the microenvironmental pH inside the formulation to help in the dissolution of the acidic enteric polymer capsule shell [23,[48][49][50][51]. In addition, sodium starch glycolate was used as the disintegrant in the capsule formulations.
The dissolution tests in FaSSIF medium were conducted to determine the effect of encapsulation on the performance of the niclosamide ASD granules ( Figure 4A). The results confirmed niclosamide concentrations of 476.0 ± 22.9 µ g/mL and 150. 7 ± 9.4 µ g/mL after 2 h of dissolution for the granules filled into Capsuline ® and Eudracap TM capsules, respectively. The Eudracaps TM took longer to dissolve, and this appeared visually to have caused the polymer gelling that resulted in a nondisintegrating ASD plug. For comparison, the extruded and milled granules without additional excipients reached niclosamide concentrations of 481.7 ± 22.2 µ g/mL after 2 h of dissolution in FaSSIF. Subsequently, enteric capsules containing niclosamide ASD were tested using pH-shift dissolution tests. The performance of the Eudracap™ was similar but more variable (168.7 ± 73.8 µ g/mL) when compared to its performance in the FaSSIF media alone. Surprisingly, the performance of the Capsuline ® was greatly reduced to 24.5 ± 2.3 µ g/mL due to the premature rupture of the capsule shells in the acidic media ( Figure 4B). The 2 h acid phase of this dissolution test was repeated, and the capsules were removed from the dissolution vessel and dried to conduct WAXS. The WAXS revealed that niclosamide in the ASD had crystallized into niclosamide hydrate (Ha) while contained within the Capsuline ® capsules. In contrast, Eudracaps TM maintained amorphous niclosamide ASD without crystallization ( Figure 4C). An acid uptake test was performed, as shown in Figure 4D. As described in the table, Capsuline ® capsules almost doubled their weight due to the rupture of the The niclosamide ASD granules were encapsulated in two different commercially available size 0 capsules: Capsuline ® (HPMCP/HPMC) and Eudracap TM enteric (HPMC/Eudragit ® polymers) capsules. These enteric capsules contain weakly acidic polymers that release their contents in the intestines, where the polymer is soluble and the capsule shell dissolves to release its payload [47]. In addition, it has been reported that PVP-VA gelling, which is contained in the ASD composition, hinders the disintegration of the dosage forms and their drug release. Gelling can be delayed by adding inorganic salts into the formulation [23]. Therefore, sodium bicarbonate was added for two reasons. First, the bicarbonate ion is a kosmotrope and delays the polymer gelling, and second, it increases the microenvironmental pH inside the formulation to help in the dissolution of the acidic enteric polymer capsule shell [23,[48][49][50][51]. In addition, sodium starch glycolate was used as the disintegrant in the capsule formulations.
The dissolution tests in FaSSIF medium were conducted to determine the effect of encapsulation on the performance of the niclosamide ASD granules ( Figure 4A). The results confirmed niclosamide concentrations of 476.0 ± 22.9 µg/mL and 150. 7 ± 9.4 µg/mL after 2 h of dissolution for the granules filled into Capsuline ® and Eudracap TM capsules, respectively. The Eudracaps TM took longer to dissolve, and this appeared visually to have caused the polymer gelling that resulted in a nondisintegrating ASD plug. For comparison, the extruded and milled granules without additional excipients reached niclosamide concentrations of 481.7 ± 22.2 µg/mL after 2 h of dissolution in FaSSIF. Subsequently, enteric capsules containing niclosamide ASD were tested using pH-shift dissolution tests. The performance of the Eudracap™ was similar but more variable (168.7 ± 73.8 µg/mL) when compared to its performance in the FaSSIF media alone. Surprisingly, the performance of the Capsuline ® was greatly reduced to 24.5 ± 2.3 µg/mL due to the premature rupture of the capsule shells in the acidic media ( Figure 4B). The 2 h acid phase of this dissolution test was repeated, and the capsules were removed from the dissolution vessel and dried to conduct WAXS. The WAXS revealed that niclosamide in the ASD had crystallized into niclosamide hydrate (Ha) while contained within the Capsuline ® capsules. In contrast, Eudracaps TM maintained amorphous niclosamide ASD without crystallization ( Figure 4C). An acid uptake test was performed, as shown in Figure 4D. As described in the table, Capsuline ® capsules almost doubled their weight due to the rupture of the capsule shell wall. In contrast, the weight gain of the Eudracaps TM was lower after the acid uptake test ( Figure 4D). Moreover, the Capsugel ® V-Caps ® Enteric size 00 capsules were tested and exhibited similar results to the Capsuline ® capsules. Based on these results, enteric capsules for use with niclosamide ASD were not considered feasible, thus enteric-coated tablets were developed. Several authors have reported similar results with enteric capsules. capsule shell wall. In contrast, the weight gain of the Eudracaps TM was lower after the acid uptake test ( Figure 4D). Moreover, the Capsugel ® V-Caps ® Enteric size 00 capsules were tested and exhibited similar results to the Capsuline ® capsules. Based on these results, enteric capsules for use with niclosamide ASD were not considered feasible, thus entericcoated tablets were developed. Several authors have reported similar results with enteric capsules. Fu et al. (2020) and Al-Tabakha et al. (2014) observed that DRcaps ® Enteric (HPMC/gellan gum) from Capsugel ruptured in acidic conditions [52,53]. In addition, Al-Tabakha et al. (2014) also observed capsule rupture with the AR-CAPS ® (HPMC/HPMCP) from CapsCanada ® .

Niclosamide ASD Was Protected by Enteric-Coated Tablet Formulation
To overcome the disadvantages of the enteric capsules, enteric-coated tablets were studied. In general, tablets are a preferred oral dosage form because of their portability, manufacturability, stability, and ease of administration, higher dose precision, and scalability [54,55]. Tablets, as monolithic (i.e., single unit tablet) oral dosage forms, have the advantage of requiring lower amounts of enteric coatings compared to multiparticulate systems, as reported by Missaghi et al. (2010) [56]. In their work related to protecting tablets containing proton pump inhibitors, a 14% weight gain of Acryl-EZE ® sufficiently coated the tablets and provided enteric protection. In contrast, multiparticulate systems required weight gains from 26% to even 61.5% of enteric coating [56].
The excipients used in the preparation of niclosamide ASD containing tablets is different from those employed in the capsule formulation. NaCl was incorporated as a kosmotropic salt to facilitate the dissolution of the tablets by reducing PVP-VA gelling [23]. After testing different disintegrants, it was determined that the most suitable disintegrant was croscarmellose sodium. In addition, microcrystalline cellulose was added to increase the hardness and reduce the friability of the tablets to withstand the coating process. The

Niclosamide ASD Was Protected by Enteric-Coated Tablet Formulation
To overcome the disadvantages of the enteric capsules, enteric-coated tablets were studied. In general, tablets are a preferred oral dosage form because of their portability, manufacturability, stability, and ease of administration, higher dose precision, and scalability [54,55]. Tablets, as monolithic (i.e., single unit tablet) oral dosage forms, have the advantage of requiring lower amounts of enteric coatings compared to multiparticulate systems, as reported by Missaghi et al. (2010) [56]. In their work related to protecting tablets containing proton pump inhibitors, a 14% weight gain of Acryl-EZE ® sufficiently coated the tablets and provided enteric protection. In contrast, multiparticulate systems required weight gains from 26% to even 61.5% of enteric coating [56].
The excipients used in the preparation of niclosamide ASD containing tablets is different from those employed in the capsule formulation. NaCl was incorporated as a kosmotropic salt to facilitate the dissolution of the tablets by reducing PVP-VA gelling [23]. After testing different disintegrants, it was determined that the most suitable disintegrant was croscarmellose sodium. In addition, microcrystalline cellulose was added to increase the hardness and reduce the friability of the tablets to withstand the coating process. The poly (vinyl alcohol)-based coating known as Opadry ® II Clear was first used as a seal coating for the tablets using water or ethanol/water mixtures. Opadry ® II is a water-soluble and pH-independent coating used for immediate release applications [57]. It was used to reduce the moisture penetration and tablet friability during the subsequent coating steps and to reduce the interactions of the enteric coating with the tablet core. After application of the seal coating, the tablets were enteric coated with Acryl-EZE ® 93 A (water) or Opadry ® Enteric (ethanol/water).
The enteric-coated tablets containing niclosamide ASD underwent the acid uptake test, and their percentages of weight gain were noticeably lower than those obtained by the enteric capsules (Table 1). Then, the tablets were tested by the dissolution test, as shown in Figure 5. The Opadry II Clear (ethanol/water) overcoated with the Opadry Enteric (ethanol/water) coating noticeably hindered the performance of the ASD formulation. In general, aqueous-based coatings are preferred over organic solvent-based coatings due to the potential toxicity, environmental, and manufacturing safety considerations [56]. In general, the aqueous-based coating formulation achieved better results, maintaining the dissolution performance of niclosamide ASD granules. These tablets were then tested in beagle dogs. and pH-independent coating used for immediate release applications [57]. It was used to reduce the moisture penetration and tablet friability during the subsequent coating steps and to reduce the interactions of the enteric coating with the tablet core. After application of the seal coating, the tablets were enteric coated with Acryl-EZE ® 93 A (water) or Opadry ® Enteric (ethanol/water). The enteric-coated tablets containing niclosamide ASD underwent the acid uptake test, and their percentages of weight gain were noticeably lower than those obtained by the enteric capsules (Table 1). Then, the tablets were tested by the dissolution test, as shown in Figure 5. The Opadry II Clear (ethanol/water) overcoated with the Opadry Enteric (ethanol/water) coating noticeably hindered the performance of the ASD formulation. In general, aqueous-based coatings are preferred over organic solvent-based coatings due to the potential toxicity, environmental, and manufacturing safety considerations [56]. In general, the aqueous-based coating formulation achieved better results, maintaining the dissolution performance of niclosamide ASD granules. These tablets were then tested in beagle dogs.  The aqueous-based enteric-coated tablets were administered to fasted beagle dogs, as shown in Figure 6. Plasma concentrations of up to 149 ± 79.2 ng/mL were detected at an administered dose of 75 mg/kg niclosamide (individual results are shown in Figure  S3). These plasma concentrations of niclosamide were higher than those previously reported in the literature using solubilized niclosamide administered at a higher dose (100 The aqueous-based enteric-coated tablets were administered to fasted beagle dogs, as shown in Figure 6. Plasma concentrations of up to 149 ± 79.2 ng/mL were detected at an administered dose of 75 mg/kg niclosamide (individual results are shown in Figure S3). These plasma concentrations of niclosamide were higher than those previously reported in the literature using solubilized niclosamide administered at a higher dose (100 mg/kg) to beagle dogs [58]. In Choi et al. (2021), niclosamide was dissolved in a mixture containing solvents, polymers, and pH modifiers (DMSO, PEG, NaOH, and saline buffer) [58]. This demonstrates that the enteric-coated tablets can protect niclosamide ASD from the gastric environment, thus preventing its crystallization in acidic conditions. Pharmaceutics 2022, 14, x FOR PEER REVIEW 11 of 15 mg/kg) to beagle dogs [58]. In Choi et al. (2021), niclosamide was dissolved in a mixture containing solvents, polymers, and pH modifiers (DMSO, PEG, NaOH, and saline buffer) [58]. This demonstrates that the enteric-coated tablets can protect niclosamide ASD from the gastric environment, thus preventing its crystallization in acidic conditions.  2021) were added to the chart as a comparison [58]. The data were extracted from the publication using the software PlotDigitizer. Reprinted/adapted with permission from Ref. [58]. 2022, Elsevier"

Conclusions
We report the development of an enteric-coated tablet dosage form of niclosamide ASD. Niclosamide ASD is an amorphous solid dispersion that generates amorphous nanoparticles during its dissolution in FaSSIF. To maximize niclosamide's oral bioavailability, the ASD cannot be exposed to acidic conditions due to the fact of API recrystallization and inferior dissolution. To overcome this acidic pH limitation, commercially available enteric capsules were tested. However, their performance was unsatisfactory for a formulation such as niclosamide ASD. Among the tested enteric capsules, acid penetration into the enteric capsules and recrystallization of the API were observed. Additionally, some capsules took too long to disintegrate, leading to polymer gelling and the formation of a nondispersible plug of the ASD within the capsule, negating the solubility advantages of the ASD granules. Therefore, enteric-coated tablets were developed and tested after experiencing the mentioned drawbacks with enteric capsules. The enteric-coated tablets successfully protected the formulation from acid and avoided polymer gelling during the dissolution tests. When tested in a beagle dog model, the enteric-coated tablets achieved increased niclosamide plasma levels when compared to a previously reported oral organic solution of niclosamide. These results will support further in vivo testing of niclosamide ASD.
Supplementary Materials: The following supporting information can be downloaded at: www.mdpi.com/xxx/s1, Figure S1: Dissolution profile of niclosamide ASD granules in buffer pH 6.5 and FaSSIF media; Figure S2 2021) were added to the chart as a comparison [58]. The data were extracted from the publication using the software PlotDigitizer. Reprinted/adapted with permission from Ref. [58]. 2022, Elsevier.

Conclusions
We report the development of an enteric-coated tablet dosage form of niclosamide ASD. Niclosamide ASD is an amorphous solid dispersion that generates amorphous nanoparticles during its dissolution in FaSSIF. To maximize niclosamide's oral bioavailability, the ASD cannot be exposed to acidic conditions due to the fact of API recrystallization and inferior dissolution. To overcome this acidic pH limitation, commercially available enteric capsules were tested. However, their performance was unsatisfactory for a formulation such as niclosamide ASD. Among the tested enteric capsules, acid penetration into the enteric capsules and recrystallization of the API were observed. Additionally, some capsules took too long to disintegrate, leading to polymer gelling and the formation of a nondispersible plug of the ASD within the capsule, negating the solubility advantages of the ASD granules. Therefore, enteric-coated tablets were developed and tested after experiencing the mentioned drawbacks with enteric capsules. The enteric-coated tablets successfully protected the formulation from acid and avoided polymer gelling during the dissolution tests. When tested in a beagle dog model, the enteric-coated tablets achieved increased niclosamide plasma levels when compared to a previously reported oral organic solution of niclosamide. These results will support further in vivo testing of niclosamide ASD.