Relevance of Liquid-Liquid Phase Separation of Supersaturated Solution in Oral Absorption of Albendazole from Amorphous Solid Dispersions

pharmaceutics

Article Relevance of Liquid-Liquid Phase Separation of Supersaturated Solution in Oral Absorption of Albendazole from Amorphous Solid Dispersions

Kyosuke Suzuki 1,*, Kohsaku Kawakami 2,* , Masafumi Fukiage 3, Michinori Oikawa 4, Yohei Nishida 5, Maki Matsuda 6 and Takuya Fujita 7

1 Pharmaceutical and ADMET Research Department, Daiichi Sankyo RD Novare Co., Ltd., 1-16-13, Kitakasai, Edogawa-ku, Tokyo 134-8630, Japan 2 Research Center for Functionals Materials, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan 3 Pharmaceutical R&D, Ono Pharmaceutical Co., Ltd., 3-3-1, Sakurai, Shimamoto-cho, Mishima-gun, Osaka 618-8585, Japan; [email protected] 4 Pharmaceutical Development Department, Sawai Pharmaceutical Co., Ltd., 5-2-30, Miyahara, Yodogawa-ku, Osaka 532-0003, Japan; [email protected] 5 Technology Research & Development, Sumitomo Dainippon Pharma Co., Ltd., 33-94, Enoki-cho, Suita, Osaka 564-0053, Japan; [email protected] 6 Research & Development Division, Towa Pharmaceutical Co., Ltd., 134, Chudoji Minami-machi, Shimogyo-ku, Kyoto 600-8813, Japan; [email protected] 7 College of Pharmaceutical Sciences, Ritsumeikan University, 1-1-1 Noji-Higashi, Kusatsu, Shiga 525-8577, Japan; [email protected] * Correspondence: [email protected] (K.S.); [email protected] (K.K.); Tel.: +81-80-4383-5853 (K.S.); +81-29-860-4424 (K.K.) Citation: Suzuki, K.; Kawakami, K.; Fukiage, M.; Oikawa, M.; Nishida, Y.; Abstract: Amorphous solid dispersion (ASD) is one of the most promising formulation technologies Matsuda, M.; Fujita, T. Relevance of for improving the oral absorption of poorly soluble drugs, where the maintenance of supersaturation Liquid-Liquid Phase Separation of plays a key role in enhancing the absorption process. However, quantitative prediction of oral Supersaturated Solution in Oral Absorption of Albendazole from absorption from ASDs is still difficult. Supersaturated solutions can cause liquid-liquid phase sepa- Amorphous Solid Dispersions. ration through the spinodal decomposition mechanism, which must be adequately comprehended Pharmaceutics 2021, 13, 220. to understand the oral absorption of drugs quantitatively. In this study, albendazole (ALZ) was https://doi.org/10.3390/ formulated into ASDs using three types of polymers, poly(methacrylic acid-co-methyl methacrylate) pharmaceutics13020220 (Eudragit) L100, Vinylpyrrolidone-vinyl acetate copolymer (PVPVA), and hydroxypropyl methylcellulose acetate succinate (HPMCAS). The oral absorption of ALZ in rats administered as ASD Academic Editor: Duncan Craig suspensions was not explained by dissolution study but was predicted using liquid-liquid phase Received: 25 December 2020 separation concentration, which suggested that the absorption of ALZ was solubility-limited. The Accepted: 25 January 2021 oral administration study in dogs performed using solid capsules demonstrated the low efficacy Published: 5 February 2021 of ASDs because the absorption was likely to be limited by dissolution rate, which indicated the importance of designing the final dosage form of the ASDs. Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in Keywords: amorphous solid dispersions; albendazole; oral absorption; liquid-liquid phase separa- published maps and institutional affil- iations. tion; dissolution study; supersaturation

1. Introduction Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. Candidate compounds developed in the pharmaceutical industry frequently exhibit This article is an open access article extremely low aqueous solubility and may be poorly absorbed after oral administration, distributed under the terms and which is one of the major issues in drug development. For such compounds, supersaturat- conditions of the Creative Commons ing formulation technologies are often used to improve their oral absorption [1–3]. The Attribution (CC BY) license (https:// method for predicting drugs’ oral absorption from supersaturating formulations using creativecommons.org/licenses/by/ dissolution studies is still under debate [4]. Speciﬁcally, non-sink conditions must be used 4.0/).

Pharmaceutics 2021, 13, 220. https://doi.org/10.3390/pharmaceutics13020220 https://www.mdpi.com/journal/pharmaceutics Pharmaceutics 2021, 13, x FOR PEER REVIEW 2 of 13

in the analysis because supersaturation must be investigated during the dissolution process [5]. However, the dissolution pattern of compounds in supersaturating formulations is significantly influenced by supersaturation degree [6,7]. The pH shift from acidic to neutral conditions is worth being tested, especially when the drugs, excipients, or both have pH-dependent dissolution properties [8]. The absorption sink’s role also needs to be considered because it changes the degree of supersaturation [6]. The effects of surface- active agents, including drug solubilization [9,10] and acceleration of crystallization [8], have complicated effects on the supersaturating dissolution behavior, which must also be considered. On the other hand, the dissolution test should be simple to ensure intra-individual reproducibility. Pharmaceutics 2021, 13, 220 2 of 13 Amorphous solid dispersion (ASD) is one of the most promising techniques for improving poorly soluble candidates’ solubility. Dissolution profiles of ASDs are quite dif- ferentin the analysis from conventional because supersaturation formulations must beusing investigated crystalline during drugs, the dissolution characterized pro- by supersat- urationcess [5]. However, and a gradual the dissolution decrease pattern in ofconcentration compounds in supersaturating[1,11]. During formulations this process, is the supersat- uratedsignificantly solution influenced may by cause supersaturation liquid-liquid degree phase [6,7]. The separation pH shift from (LLPS) acidic tobased neutral on the spinodal decompositionconditions is worth being mechanism tested, especially [11,12]. when This the drugs, process excipients, produces or both have highly pH- concentrated dependent dissolution properties [8]. The absorption sink’s role also needs to be considered (nano)becausedroplets/particles, it changes the degree ofwhich supersaturation cannot be [6 ].absorbed The effects directly of surface-active but may agents, act as drug reservoirs,including shuttles, drug solubilization or both to [9 ,10carry] and drug acceleration molecules of crystallization effectively [8], in have the complicated mucus layer [9,13–15]. Theeffects drug on the concentration supersaturating in dissolution the continuum behavior, phase which must(i.e., alsoLLPS be considered.concentration) On the was relevant to oralother absorption hand, the dissolution [4,8]. Thus, test should it is critical be simple to to understand ensure intra-individual and control reproducibility. LLPS after the dissolution Amorphousof ASDs tosolid use the dispersion ASD technology (ASD) is one effectivel of the mosty. promising techniques for improving poorly soluble candidates’ solubility. Dissolution profiles of ASDs are quite differentAlbendazole from conventional (ALZ) formulationsis a weak basic using compound crystalline drugs,with low characterized aqueous bysolubility su- (Figure 1, Tablepersaturation 1) [16– and18]. a Some gradual vari decreaseability in can concentration be found [1for,11 ].its During reported this process,intrinsic the solubility, but it issupersaturated typically below solution 1 μ mayg/mL, cause which liquid-liquid is slightly phase improved separation in (LLPS)fasted based-state onsimulated the intestinal fluidspinodal (FaSSIF) decomposition and significantly mechanism [11enhanced,12]. This processin fed- producesstate simulated highly concentrated intestinal fluid (FeSSIF). (nano)droplets/particles, which cannot be absorbed directly but may act as drug reservoirs, Althoshuttles,ugh or bothsome to successful carry drug moleculesattempts effectivelyto apply inASD the mucusto ALZ layer to improve [9,13–15]. its The oral absorption havedrug concentration already been in thereported continuum, the phase detailed (i.e.,LLPS mechanism concentration) of the was enhanced relevant to oral oral absorption has notabsorption yet been [4,8 ].elucidated. Thus, it is critical Kohri to et understand al. [19] showed and control that LLPS an afterASD the of dissolutionALZ with of hydroxypropyl mASDsethylcellulose to use the ASD (HPMC) technology and effectively. HPMC phthalate significantly improved the dissolution propertiesAlbendazole and (ALZ)maintained is a weak supersaturatio basic compoundn withfor > low8 h. aqueous In the oral solubility administration (Figure1, study of an Table1)[ 16–18]. Some variability can be found for its reported intrinsic solubility, but it is ALZtypically ASD below in rabbits 1 µg/mL, with which low is slightlygastric improvedacidity, the in fasted-state area under simulated the concentration intestinal -time curve (AUC)fluid (FaSSIF) was three and significantly-fold higher enhanced than that in fed-state of the physical simulated mixture. intestinal fluidDespite (FeSSIF). the successful ap- plicationAlthough someof ASD successful in this attempts animal to applystudy, ASD no toinsights ALZ to improveabout the its oral formulation absorption strategy were obtainedhave already because been reported, only one the detailedASD was mechanism investigated of the enhanced in their oralstudy. absorption Silvina has et al. [20] found not yet been elucidated. Kohri et al. [19] showed that an ASD of ALZ with hydroxypropyl thatmethylcellulose the AUC (HPMC)after oral and administration HPMC phthalate of significantly an ALZ ASD improved to mice the dissolutionusing Pluronic 188 was muchproperties higher and maintainedthan that that supersaturation after administration for >8 h. In theof its oral crystalline administration suspension study of [20]. The ASD exhibitedan ALZ ASD a superior in rabbits withdissolution low gastric rate acidity, compared the area to under that of the the concentration-time crystalline ALZ. In this case, thecurve improvement (AUC) was three-fold in dissolution higher than behavior that of the and physical oral mixture. absorption Despite was the mainly successful attributed to the application of ASD in this animal study, no insights about the formulation strategy were polymer’s surfactant-like properties, including enhancement of wettability and the micel- obtained because only one ASD was investigated in their study. Silvina et al. [20] found larthat solubilizat the AUC afterion oral effect. administration of an ALZ ASD to mice using Pluronic 188 was muchEarlier higher than studies that that of ALZ after administration ASDs did not of itsrecognize crystalline the suspension occurrence [20]. Theof LLPS ASD during dissolution.exhibited In a this superior study, dissolution the LLPS rate behavior compared of to supersaturated that of the crystalline ALZ ALZ. solutions In this case, was investigated, andthe improvement then an oral in dissolutionadministration behavior study and oralof the absorption ASDs waswas mainly conducted attributed in rats to and dogs to the polymer’s surfactant-like properties, including enhancement of wettability and the observemicellar solubilization the relevance effect. of LLPS behavior to oral absorption.

Figure 1. 1.Chemical Chemical structure structure of ALZ. of ALZ. Earlier studies of ALZ ASDs did not recognize the occurrence of LLPS during dissolution. In this study, the LLPS behavior of supersaturated ALZ solutions was investigated, and then an oral administration study of the ASDs was conducted in rats and dogs to observe the relevance of LLPS behavior to oral absorption.

Pharmaceutics 2021, 13, 220 3 of 13

Table 1. Basic Properties of ALZ [16–18].

Molecular Weight pKa (Base), 25 ◦C LogP, 25 ◦C 265.3 4.2 3.1 Solubility, 37 ◦C(µg/mL) FeSSIF FeSSIF FaSSIF FaSSIF pH 1.2 blk blk pH 7.4 (pH 5.0) (pH 5.0) (pH 6.5) (pH 6.5) 184 1.1 6.1 0.85 1.9 0.75 Subscript “blk” means the removal of taurocholic acid and lecithin from simulated intestinal ﬂuids.

2. Materials and Methods 2.1. Chemicals ALZ was purchased from Sigma-Aldrich (St. Louis, MO, USA) and was ground using a mortar and pestle before use. Poly(methacrylic acid-co-methyl methacrylate) (Eudragit L100), hydroxypropyl methylcellulose acetate succinate (HPMCAS), and vinylpyrrolidone- vinyl acetate copolymer (Kollidon VA64, PVPVA) were obtained from Evonik Industries AG (Essen, Germany), Shin-Etsu Chemical Co., Ltd. (Tokyo, Japan), and BASF SE (Lud- wigshafen am Rhein, Germany), respectively. Mannitol was supplied by Nacalai Tesque, Inc. (Kyoto, Japan). All chemicals were used as supplied.

2.2. Physical Characterization Powder X-ray diffraction (PXRD) measurements were performed using a D8 DIS- COVER with a general area detector diffraction system (Bruker, MA, USA) using CuKα radiation. The tube voltage and current were 40 kV and 40 mA, respectively. Data were collected in the range of 5◦ to 40◦ with intervals of 0.02◦ (2 theta), and the scan speed was 5◦/min. Simultaneous measurement of thermogravimetric (TG) and differential thermal analysis (DTA) was performed on TG/DTA 6200 (Hitachi High-Tech, Tokyo, Japan) with a scan rate of 10 ◦C/min. The temperature was calibrated with indium. Approximately 2 mg of the sample (as an ALZ equivalent) was evaluated using an aluminum open pan. Dry nitrogen was used as an inert gas at a ﬂow rate of 100 mL/min. The formulation morphology was examined using scanning electron microscopy (SEM) with the SU8000 SEM (Hitachi, Tokyo, Japan) at an accelerating voltage of 1 kV. Samples were loaded on carbon tapes under nitrogen ﬂow to avoid moisture adsorp- tion, followed by sputter coating using a platinum coater (E-1030 ion sputter, Hitachi) before analysis.

2.3. Solubility Measurement of ALZ Approximately 10 mg of crystalline ALZ was loaded in a test tube (n = 3), to which 5 mL of phosphate buffer (PB, 50 mM, pH 7) was added in the presence or absence of the polymer (0.1 w/v%). The solutions were then lightly sonicated, vortexed, and then rotated at approximately 50 rpm at 25 or 37 ◦C for 1 day. Each solution was filtered using a nylon syringe filter with a pore size of 0.2 µm (Sanplatec, Osaka, Japan). For the measurement of solubility at 37 ◦C, the syringes and syringe filters were warmed in an oven at 37 ◦C prior to use. The filtrate was diluted with ethanol and measured using a high-performance liquid chromatography (HPLC) system equipped with an octadecylsilyl (ODS) YMC-Pack Pro C18 column (150 mm × 2.0 mm ID, YMC, Kyoto, Japan). A water/acetonitrile mixture at a ratio of 55:45 was used as the mobile phase at a flow rate of 0.2 mL/min. The injection volume and the column temperature were 2 mL and 30 ◦C, respectively. Measurements were made at a concentration range from 0.1 to 20 µg/mL at a wavelength of 295 nm using a photodiode array detector, where the linearity was confirmed.

2.4. Preparation of ASDs of ALZ ALZ and the polymer were dissolved separately in 1,4-dioxane at concentrations of 10 and 20 mg/mL, respectively, and then the solutions were mixed to achieve an Pharmaceutics 2021, 13, 220 4 of 13

ALZ:polymer ratio of 1:3 (w/w). For the PVPVA, a solution with an ALZ:polymer ratio of 1:4 was also prepared. The solutions were frozen using liquid nitrogen and then freeze- dried using the VirTis Advantage EL freeze dryer (Warminster, PA, USA). The shelf was initially maintained at −50 ◦C, followed by successive drying at −20 ◦C, −5 ◦C, and 10 ◦C for 10 h each. Finally, the residual solvents were removed at 35 ◦C under vacuum, and each ASD was stored in a refrigerator before use. An equal amount of crystalline ALZ and mannitol were mixed using a mortar and pestle to prepare a physical mixture as the control sample used for the dissolution and administration studies. All the formulations were conﬁrmed to be physically and chemically stable during the storage (for at least three months).

2.5. LLPS Concentration and Particle Properties An appropriate amount of a dimethylacetamide solution of ALZ was added to 50 mL PB in the polymer’s presence or absence (0.1 w/v%). The solutions were stirred at approximately 200 rpm at 25 ◦C. The turbidity was measured at a wavelength of 500 nm using a DU-800 spectrophotometer (Beckman Coulter, Brea, CA, USA) after stirring for 30 min. The phase separation concentration was estimated from the breakpoint of the turbidity-concentration curves [7,8]. The suspensions’ particle size and zeta potential were measured using a Zetasizer Nano ZS (Malvern Panalytical, Malvern, UK). All measurements were repeated three times, and the mean particle size was determined using the cumulant method. LLPS behavior was also observed using ASD as a starting material. ASD or crystalline ALZ (as a physical mixture with mannitol) was dispersed at a concentration of 40 µg/mL as the ALZ equivalent in 50 mL PB at 25 ◦C while stirring at approximately 200 rpm in beakers. Solutions were sampled over time and immediately filtered through syringe filters with a pore size of 0.2 or 0.7 µm (Millex-LG, Merck Millipore, Burlington, MA, USA and GF/F, Whatman, Buckinghamshire, UK, respectively). The ALZ concentration of the filtered solutions was determined using HPLC as described above. The measurements were duplicated to obtain averaged values. This observation is described as the LLPS dissolution test hereafter.

2.6. µDISS Dissolution Study The dissolution study was performed using a µDISS Profiler™ system (pION Inc., Billerica, MA, USA). The first and second fluids of the Japanese Pharmacopeia (JP1 and JP2, respectively) were used as dissolution media. The µDISS vessels were placed in a thermostatic chamber at 37 ◦C, 10 mL of JP1 (pH 1.2) or JP2 (pH 6.8) was added to the vessel, and then the UV probe was immersed in the solution. The ASD or crystalline ALZ was dispersed at a concentration of 15 µg/mL as the ALZ equivalent while stirring at 200 rpm, and the UV spectra were acquired over time to measure the ALZ concentration.

2.7. Oral Administration Study in Rats All experiments using rats were approved by the Ethical Review Committee of Dai- ichiSankyo RD Novare (Exp. No. 2017-024, 2017-027 (2017)). Male Crl:CD(SD) rats (6–7-week-old, Charles River Laboratories Japan, Yokohama, Japan) were housed in a temperature-controlled room at 23 ± 2 ◦C with a relative humidity of 55 ± 20%, and a 12 h light/dark cycle. The rats were starved for 16 h before and 6 h after administration with free access to water. Formulations were dispersed in 0.5% methylcellulose (MC) (for crystalline ALZ) or puriﬁed water (for ASDs) using a homogenizer, followed by imme- diate administration to rats (n = 3) at a dose of 10 mg/10 mL/kg as the ALZ equivalent. Approximately 150 µL blood samples were withdrawn through the jugular vein using pre-heparinized syringes 0.25, 0.5, 1, 2, 4, and 8 h after administration. Plasma samples were prepared by centrifuging the blood at 3000× g for 3 min and were then stored at −20 ◦C until the analysis. Pharmaceutics 2021, 13, 220 5 of 13

Plasma concentration of ALZ sulfoxide, a metabolite of ALZ, was measured using a liquid chromatography-tandem mass spectrometry (LC-MS/MS) method (API-4000, SCIEX UPLC, Waters, Milford, MA, USA). The detailed procedure was described elsewhere [7]. Briefly, the plasma sample (20 µL) was mixed with a 50% (v/v) aqueous acetonitrile solution, to which 200 µL of acetonitrile/methanol (75/25 (v/v)) was added. Niflumic acid was used as the internal standard. The mixture was filtered and measured. The linearity was confirmed in a concentration range of 0.005–5 µg/mL.

2.8. Oral Administration Study in Dogs All experiments in the beagle dogs were approved by the Ethical Review Committee of KAC Co. (Exp. No. 17-1220 (2017)). Male beagle dogs (body weight: 10.8–13.2 kg, Kitayama Labes, Ina, Japan) were housed in a temperature-controlled room at 23–27 ◦C under a 12 h light/dark cycle. The dogs were starved 16 h before and for 4 h after administration and allowed free access to water. Each formulation was loaded into #00 gelatin capsules, except for the HPMCAS ASD, for which aqueous suspension was also prepared, and then administered to the dogs (n = 3) at a dose of 10 mg/0.5 mL/kg. Pentagastrin was administered intramuscularly at a dose of 10 µg/kg 30 min before and 30/90 min after administration. Approximately 2.5 mL blood samples were withdrawn using pre-heparinized syringes 0.25, 0.5, 1, 2, 4, and 7 h after administration, centrifuged at 1800× g for 10 min and then stored at −80 ◦C until the analysis. The plasma ALZ sulfoxide concentration was measured using LC-MS/MS (Ultimate 3000 Rapid Separation LC, Q Exactive, Thermo Scientific, Waltham, MA, USA). The plasma samples were processed in the same manner as done for the rat study with phenacetin as the internal standard. The mixture was filtered and subjected to measurements. Linearity was confirmed in a concentration range of 0.0003–1 µg/mL.

3. Results 3.1. Physicochemical Properties of Crystalline ALZ and ASDs The PXRD and TG-DTA patterns of crystalline ALZ are presented in Figure2a,b, respectively. Sharp diffraction peaks were confirmed in the PXRD pattern. An endothermic melting peak was found at 199 ◦C in the DTA curve; however, thermal degradation was also initiated approximately at the same temperature. Thus, thermal analysis was not an appropriate method for analyzing the crystalline property of ALZ because of the overlapping of melting and degradation behaviors. No crystalline diffraction peaks were observed for the ASDs prepared using Eudragit or HPMCAS (Figure2c), whereas the PVPVA ASD showed small diffraction peaks at an ALZ:polymer ratio of 1:3. ALZ Crystals were found only for this ASD in polarized microscopy (PLM) analysis (supplementary material). The diffraction peak disappeared when the PVPVA content was increased to a ratio of 1:4 (Figure2c), which was also confirmed in the PLM image (supplementary material). Thus, this ratio was employed only for PVPVA in the following studies. Figure2d shows the TG-DTA pattern of PVPVA/ALZ = 3/1 ASD. Although a small broad endothermic peak was found at the melting temperature in the DTA curve, it was difficult to assign this event as the melting because of decrease of the weight at the same temperature that indicated degradation. Thus, thermal analysis was not reliable for finding a small amount of crystalline ALZ. Nevertheless, no melting behaviors were observed for all ASDs (supplementary material). Absence of ALZ crystals was also confirmed by the PLM analysis (supplementary material). Figure3 shows SEM images of the crystalline ALZ and its ASDs. The size of the ALZ crystals was confirmed to be reduced to a micrometer order after grinding. A porous structure was confirmed for all the ASDs, which is typical for freeze-dried ASDs, and no crystalline particles were found. Pharmaceutics 2021 13 Pharmaceutics 2021, 13, x FOR PEER, REVIEW, 220 6 of 13 6 of 13 Pharmaceutics 2021, 13, x FOR PEER REVIEW 6 of 13

100 100

99 )

) 99

(

)

) o

h 98

(

o h

98 A

(

G T

97 A

D G

T 97 96 5 mV/g 96 5 mV/g 5 10 20 30 40 95 5 10 20 30 40 20 60 91500 140 180 220 T2e0mperature6 0(oC) 100 140 180 220 o (a) (b) Temperature ( C) (a) 100 (b) 100

98 98

y A

)

p (

s 96

y t

)

) o

i B

(

i s

t 96

(

n h

B o

C1 T t

94 E

(

C1 94 T

C2 G

D T C92 1 mV/g 92 5 10 20 30 40 1 mV/g 90 2 Th5eta 1(0Degree) 20 30 40 20 60 91000 140 180 220 2 Theta ( Degree) o Te2m0 perature6 (0C) 100 140 180 220 Temperature (oC) (c) (d) (c) (d) Figure 2. (a) PXRDFigure pattern 2. (a) PXRD of crystalline pattern of ALZ. crystalline (b) TG-DTA ALZ. curve(b) TG of-DTA crystalline curve of ALZ. crystalline (c) PXRD ALZ. patterns (c) PXRD of ALZ pat- ASDs. A: Figure 2. (a) PXRD pattern of crystalline ALZ. (b) TG-DTA curve of crystalline ALZ. (c) PXRD pat- Eudragit/ALZterns = 3/1, of ALZ B: HPMCAS/ALZ ASDs. A: Eudragit/ALZ = 3/1, C1: = PVPVA/ALZ 3/1, B: HPMCAS/ALZ = 3/1, C2: = PVPVA/ALZ 3/1, C1: PVPVA/ALZ = 4/1. (d )= TG-DTA 3/1, C2: curve of PVPVA/ALZ terns= 4/1. of(d )ALZ TG- DTAASDs. curve A: Eudragit/ALZ of PVPVA/ALZ = 3/1,= 3/1 B: ASD. HPMCAS/ALZ = 3/1, C1: PVPVA/ALZ = 3/1, C2: PVPVA/ALZ = 3/1 ASD. PVPVA/ALZ = 4/1. (d) TG-DTA curve of PVPVA/ALZ = 3/1 ASD.

(a) (b) (a) (b)

(c) (d) (c) (d) Figure 3. SEM images of (a) crystalline ALZ, (b) HPMCAS ASD, (c) Eudragit ASD, and (d) PVPVA ASD. FigureFigure 3. 3. SEMSEM images images of (a of) crystalline (a) crystalline ALZ, ALZ,(b) HPMCAS (b) HPMCAS ASD, (c ASD,) Eudragit (c) Eudragit ASD, and ASD, (d) PVPVA and (d ) ASD.PVPVA ASD. 3.2. Dissolution and LLPS Properties of ALZ ASDs 3.2.3.2. Dissolution Dissolution and and LLPS LLPS Properties Properties of of ALZ ALZ ASDs ASDs Table 2 shows the crystalline ALZ’s equilibrium solubility, LLPS concentrations, and Table2 shows the crystalline ALZ’s equilibrium solubility, LLPS concentrations, and particle propertiesTable above 2 shows the LLPS the crystallineconcentrations. ALZ ’Thes equilibrium poor solubility solubility in PB, LLPS was slightlyconcentrations, and particle properties above the LLPS concentrations. The poor solubility in PB was slightly improved inparticle the presence properties of Eudragit above theand LLPS HPMCAS concentrations., and the solubilities The poor solubilitydid not depend in PB was slightly improved in the presence of Eudragit and HPMCAS, and the solubilities did not depend significantlyimproved on temperature in the presenceunder these of Eudragit conditions. and The HPMCAS pH did, notand change the solubilities after equili- did not depend signiﬁcantly on temperature under these conditions. The pH did not change after equi- bration in allsignificantly cases. In the onabsence temperature of the polymer, under these LLPS conditions. occurred atThe 1.4 pH μg/mL, did not which change was after equili- bration in all cases. In the absence of the polymer, LLPS occurred at 1.4 μg/mL, which was

Pharmaceutics 2021, 13, x FOR PEER REVIEW 7 of 13 Pharmaceutics 2021, 13, 220 7 of 13

>10-fold the solubility, whereas it increased in the presence of polymers in the order of 7.2 μg/mL,libration 7.0 in μg/mL, all cases. and In 3.8 the μg/mL absence for ofPVPVA, the polymer, HPMCAS, LLPS and occurred Eudragit, at 1.4 respectivµg/mL,ely. which The meanwas >10-fold particle thesize solubility, found in whereas each solution it increased was > in1 μ them, presence except for of polymersthe HPMCAS in the solution, order of ca.7.2 0.22µg/mL, μm. 7.0 µg/mL, and 3.8 µg/mL for PVPVA, HPMCAS, and Eudragit, respectively. The mean particle size found in each solution was >1 µm, except for the HPMCAS solution, Tableca. 0.22 2. Solution/suspensionµm. properties of ALZ in PB (pH7) in the presence and absence of polymers.

Table 2. Solution/suspensionPolymers properties of ALZ in PBNo (pH7) Polymer in the presenceEudragit and L100 absence H ofPMCAS polymers. PVPVA Solubility, 25 °C (μg/mL) <0.10 0.44 ± 0.02 0.47 ± 0.07 <0.10 PolymersSolubility, No Polymer 37 °C (μg/mL) Eudragit<0.10 L100 0.32 HPMCAS ± 0.06 0.55 ± 0.09 PVPVA<0.10 ◦ Solubility, 25 C(µg/mL)LLPS, <0.1025 °C (μg/mL) 0.44 ± 0.021.4 0.473.8 ± 0.077.0 <0.107.2 ◦ µ ± ± Solubility, 37 C( g/mL)Particle <0.10size, 25 °C (μm) 0.32 0.06>3 2.13 0.55 ± 0.230.09 0.22 ± 0.02 <0.101.12 ± 0.06 LLPS, 25 ◦C(µg/mL) 1.4 3.8 7.0 7.2 − Particle size, 25 ◦C(µm)Polydispersity >3 Index (PDI) 2.13 ± 0.23 0.59 0.22 ± 0.10± 0.02 0.14 ± 0.021.12 0.21± 0.06 ± 0.04 Polydispersity Index (PDI) Zeta potential,− 25 °C (mV) 0.59 ± 0.10− −32.8 0.14 ± ±1.20.02 −12.4 ± 1.10.21 −±0.20.04 ± 0.2 Zeta potential, 25 ◦C (mV) − −32.8 ± 1.2 −12.4 ± 1.1 −0.2 ± 0.2 The size distribution of the particles found in the Eudragit solution was broad, alt- houghThe they size exhibited distribution the highest of the zeta particles potential found value. in the In Eudragitthe polarized solution microscopy was broad, analy- al- sis,though crystal they particles exhibited were the highest found zetain all potential samples value. except In forthe polarized HPMCAS microscopy (data not shown). analysis, Thus,crystal the particles observed were particle found sizes in all were samples likely except not of for the HPMCAS LLPS particles (data notbut shown).were of Thus,ALZ crystals.the observed Crystallization particle sizes was were not evident likely not in the of theHPMCAS LLPS particles solution but because were the of ALZ particle crystals. size wasCrystallization too small in was the notmicroscopic evident in investigation. the HPMCAS The solution zeta potentials because the were particle highly size negative was too insmall the presence in the microscopic of Eudragit investigation. and slightly negative The zeta in potentials the presence were of highly HPMCAS, negative which in is the a typicalpresence observation of Eudragit [7,8]. and slightly negative in the presence of HPMCAS, which is a typical observationFigure 4 [a7 ,shows8]. the LLPS dissolution test results, where suspensions were treated usingFigure syringe4a filters shows with the a LLPS pore dissolutionsize of 0.7 μm. test ALZ results, concentrations where suspensions in the eluate were from treated the crystallineusing syringe ALZ ﬁlters did not with exceed a pore 0.1 size5 μg/mL, of 0.7 whichµm. ALZ was concentrationsalmost explainable in the from eluate its equi- from libriumthe crystalline solubility ALZ in did PB. notIn contrast,exceed 0.15 ALZµg/mL, concentrations which was reached almost much explainable higher from values its whenequilibrium ASDs were solubility subjected in PB. to In the contrast, study. ALZThe Eudragit concentrations ASD showed reached the much highest higher concen- values trationwhen ASDsof ALZ, were followed subjected by the to thePVPVA study. and The HPMCAS Eudragit ASDs. ASD showed The observed the highest concentrations concentra- oftion these of ALZ, ASDs followed were higher by the than PVPVA the equilibrium and HPMCAS solubility ASDs. in The the observed polymers concentrations’ presence (Ta- of blethese 2). ASDsThe concentrations were higher than of the the Eudragit equilibrium and solubility PVPVA inASDs the polymers’started to presencedecrease (Tablefrom 102). min,The concentrationsindicating that precipitation of the Eudragit was and induced PVPVA by ASDsexcess started solids or to crystallization decrease from of 10 ALZ min, orindicating both. that precipitation was induced by excess solids or crystallization of ALZ or both.

4 4 (b)

(a)

)

) L L 3

3 m

(

i t

t 2 2

n o

o 1 1

C C

0 0 0 10 20 30 0 10 20 30 Time after dissolution (min) Time after dissolution (min)

Figure 4. Concentration profiles of ALZ eluted from crystalline ALZ and ALZ ASDs in the LLPS dissolution test. Filtration Figurewas performed 4. Concentration using syringe profiles filters of ALZ with eluted pore sizesfrom ofcrystalline (a) 0.7 µm ALZ or ( band) 0.2 ALZµm. ASDs Symbols: in the Crystalline LLPS dissolution ALZ (N ),test. PVPVA Filtration ASD was performed using syringe filters with pore sizes of (a) 0.7 μm or (b) 0.2 μm. Symbols: Crystalline ALZ (◆), PVPVA (), HPMCAS ASD (N), Eudragit ASD (•). ASD (■), HPMCAS ASD (▲), Eudragit ASD (●). The LLPS dissolution tests’ filtrates using filters with 0.7 µm pore size had a slightly turbidThe appearance, LLPS dissolution which tests suggested’ filtrate thats using small filters particles with permeated0.7 μm pore through size had the a slightly syringe turbidfilters. appearance, This is especially which expected suggested for HPMCASthat small ASD particles because permeated it creates through small LLPS the particlessyringe (Table2). Therefore, the experiments were repeated using membrane filters with a 0.2 µm

Pharmaceutics 2021, 13, x FOR PEER REVIEW 8 of 13

filters. This is especially expected for HPMCAS ASD because it creates small LLPS particles (Table 2). Therefore, the experiments were repeated using membrane filters with a 0.2 μm pore size (Figure 4b). All the filtrates in this experiment were transparent. The ALZ concentrations of the HPMCAS and PVPVA ASDs decreased dramatically, indicating the presence of particles with sizes between 0.2 and 0.7 μm. The concentration of Eudragit ASD also decreased by approximately 1 μg/mL. Although the ALZ concentration after filtration with a 0.2 μm filter in the LLPS dissolution test could be expected to be equal to that of the LLPS, it was much lower. One possible explanation is enhanced precipitation induced by excess solids [7]. However, the PharmaceuticsALZ concentrations2021, 13, 220 from the HPMCAS and PVPVA ASDs agreed well with the equilib- 8 of 13 rium solubility, which was likely attributable to the crystallization of ALZ. The dissolution patterns supported this hypothesis for the Eudragit and PVPVA ASDs in Figure 4a. Crys- tallization was not suspectedpore size (Figurefor the4b). HPMCAS All the filtrates ASD from in this its experiment dissolution were pattern transparent. (Figure The ALZ 4a). The crystals mightconcentrations pass through of the the HPMCAS syringe and filters PVPVA because ASDs decreased of the dramatically,smallness of indicating the the presence of particles with sizes between 0.2 and 0.7 µm. The concentration of Eudragit particle size. Figure 5ASD shows also decreasedthe dissolution by approximately properties 1 µg/mL. of crystalline ALZ and its ASDs evaluated using the μDISSAlthough apparatus, the ALZ where concentration the order after of the filtration achieved with aconcentrations 0.2 µm filter in the did LLPS dis- not agree with that ofsolution the equilibrium test could be solubility. expected to In be the equal acidic to that enviro of the LLPS,nment, it wasthe muchbest dis- lower. One solution was observedpossible with explanationthe crystalline is enhanced ALZ and precipitation PVPVA induced ASD, followed by excess solids by HPMCAS [7]. However, the ALZ concentrations from the HPMCAS and PVPVA ASDs agreed well with the equilib- ASD and Eudragit ASD.rium None solubility, of these which formulations was likely attributable reached to equilibrium the crystallization solubility, of ALZ. Thewhich dissolution was reported to be 184patterns μg/mL supported (Table this 1), hypothesis presumably for the because Eudragit the and observ PVPVAation ASDs inwas Figure made4a. Crystal- only for 1 h. The poorlization dissolution was not suspectedbehavior for of the ALZ HPMCAS from ASDEudragit from its and dissolution HPMCAS pattern ASDs (Figure 4a). could be attributableThe to the crystalsse acidic might passpolymers through’ low the syringe solubility filters. Elution because of of the ALZ smallness was likely of the particle to be disturbed in thesize. presence Figure5 of shows these the insoluble dissolution polymers. properties of crystalline ALZ and its ASDs evaluated using the µDISS apparatus, where the order of the achieved concentrations did not agree In contrast, the dissolutionwith that of theunder equilibrium neutral solubility. pH conditions In the acidic was environment,enhanced the the most best dissolution by HPMCAS ASD, followedwas observed by PVPVA with ASD. the crystalline The concentrations ALZ and PVPVA achieved ASD, followed with these by HPMCAS ASDs ASD was higher than thatand of the Eudragit equilibrium ASD. None solubility of these formulations (Table 2), reachedindicating equilibrium that supersatura- solubility, which was tion was attained. Eudragitreported ASD to be 184alsoµ g/mLslightly (Table improved1), presumably the dissolution, because the observation where the was ALZ made only for 1 h. The poor dissolution behavior of ALZ from Eudragit and HPMCAS ASDs could concentration almostbe agreed attributable with tothe these equilibrium acidic polymers’ solubility. low solubility. Crystalline Elution ALZ of ALZ was was rarely likely to be dissolved under the neutraldisturbed pH in thecondition. presence of these insoluble polymers.

10 10

(a) Crystalline ALZ (b) Eudragit ASD

)

) L

L 8 8

( (

a r

r 4 4

c n

2 n 2

C C 0 0 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Time after dissolution(min) Time after dissolution(min) 10 10

)

) L

L 8 8

(

6 6

a r

r 4 4

c n

n 2 2

C C 0 0 0 10 20 30 40 50 60 0 10 20 30 40 50 60 Time after dissolution(min) Time after dissolution(min) FigureFigure 5. 5.µ DissμDiss dissolution dissolution test of test (a) crystallineof (a) crystalline ALZ and (ALZb–d) ALZand ASDs.(b–d) TypesALZ ofASDs. polymers Types are indicatedof polymers in the are ﬁgure. indicatedBlue and red in lines the representfigure. Blue dissolution and red proﬁles lines in represent JP1 (pH 1.2) dissolution and JP2 (pH 6.8)profiles solutions. in JP1 (pH 1.2) and JP2 (pH 6.8) solutions. In contrast, the dissolution under neutral pH conditions was enhanced the most by HPMCAS ASD, followed by PVPVA ASD. The concentrations achieved with these ASDs 3.3. Oral Administrationwas Study higher than that of the equilibrium solubility (Table2), indicating that supersaturation was attained. Eudragit ASD also slightly improved the dissolution, where the ALZ concentration almost agreed with the equilibrium solubility. Crystalline ALZ was rarely dissolved under the neutral pH condition.

Pharmaceutics 2021, 13, x FOR PEER REVIEW 9 of 13 Pharmaceutics 2021, 13, 220 9 of 13

3.3. OralFigure Administration 6 shows Studythe results of the oral absorption study of crystalline ALZ and its ASDsFigure in rats.6 shows The theabsorption results of the was oral considerably absorption study higher of crystalline with the ALZ ASDs and its relative ASDs to crystalline ALZin rats.. The The pharmacokinetic absorption was considerably (PK) parameters higher with are the shown ASDs relative in Table to crystalline 3. The PVPVA ALZ. and HPM- CASThe pharmacokineticASDs showed (PK) almost parameters the same are shown plasma in Tableconcentratio3. The PVPVAn profiles and HPMCAS with AUC values 4- ASDs showed almost the same plasma concentration proﬁles with AUC values 4-fold higher fold higher than that of the crystalline ALZ. The Eudragit ASD also improved the oral than that of the crystalline ALZ. The Eudragit ASD also improved the oral absorption; absorptionhowever, the; however, AUC was the slightly AUC lower was thanslightly those lower of the than other those two ASDs, of thewhich other wastwo ASDs, which wasapproximately approximately 3-fold of 3 the-fold crystalline of the crystalline ALZ. ALZ.

) Crystalline ALZ L

m 10 PVPVA ASD /

g HPMCAS ASD

(

n 8 Eudragit ASD

a r

t 6

c n

o 4

a m

s 2

l P 0 0 2 4 6 8 Time (h)

FiFiguregure 6.6.Plasma Plasma ALZ ALZ sulfoxide sulfoxide concentration concentration proﬁles afterprofiles oral administrationafter oral administration of crystalline ALZ of crystalline ALZ N andand its ASDsASDs in in rats rats at aat dose a dose of 10 mg/kgof 10 mg/kg (n = 3). ( Symbols:n = 3). Symbols: Crystalline Crystalline ALZ ( ), PVPVA ALZ ASD (◆), ( PVPVA), ASD (■), HPMCAS ASD ( ), Eudragit ASD (•). HPMCAS ASDN (▲), Eudragit ASD (●).

Table 3. PK parameters for oral administration study of ALZ ASDs in rats. Table 3. PK parameters for oral administration study of ALZ ASDs in rats. Formulations Crystalline ALZ Eudragit ASD HPMCAS ASD PVPVA ASD Formulations Crystalline ALZ Eudragit ASD HPMCAS ASD PVPVA ASD Cmax (µg/mL) 2.17 ± 0.54 6.30 ± 1.14 8.06 ± 0.81 8.82 ± 1.44 CmaxTmax (μ(h)g/mL) 3.3 ±2.171.2 ± 0.54 4.0 ± 0.06.30 ± 1.14 2.0 ± 0.08.06 ± 2.3 0.81± 1.5 8.82 ± 1.44 AUC (µg·hr/mL) 12.7 ± 3.5 37.4 ± 4.9 48.9 ± 5.4 52.4 ± 11.6 Tmax (h) 3.3 ± 1.2 4.0 ± 0.0 2.0 ± 0.0 2.3 ± 1.5 AUC (μg·hr/mL) 12.7 ± 3.5 37.4 ± 4.9 48.9 ± 5.4 52.4 ± 11.6 Figure7 and Table4 show the oral absorption study results of the crystalline ALZ and its ASDs in beagle dogs. Administration of HPMCAS ASD as a suspension considerably enhancedFigure the 7 oral and absorption, Table 4 similarshow tothe the oral rat study’s absorption ﬁndings. study However, results theabsorption of the crystalline ALZ andbehavior its ASDs was totally in beagle different dogs. when Administration the same ASD was of administered HPMCAS as ASD capsules, as a where suspension the consider- ablyASDs enhanced were ﬁlled asthe solid oral forms. absorption, The absorption similar from to the the HPMCAS rat study ASD’s wasfindings comparable. However, the ab- sorptionto the crystalline behavior ALZ, was which totally was different also administered when the as capsules. same ASD That was from a thedministered Eudragit as capsules, ASD was higher than that from the HPMCAS ASD, whereas that from the PVPVA ASD wherewas worse the thanASDs that were from crystallinefilled as solid ALZ. forms. The absorption from the HPMCAS ASD was comparable to the crystalline ALZ, which was also administered as capsules. That from Table 4.thePK parametersEudragit for ASD oral administrationwas higher studythan of that ALZ ASDsfrom in the beagle HPMCAS dogs. ASD, whereas that from the CrystallinePVPVA ASD wasEudragit worse than HPMCASthat from crystallineHPMCAS ALZ ASD. PVPVA Formulations ALZ ASD ASD Suspension ASD

Cmax (µg/mL) 0.30 ± 0.21 0.44 ± 0.10 0.34 ± 0.09 0.57 ± 0.05 0.20 ± 0.07 Tmax (h) 1.3 ± 0.6 1.7 ± 0.6 2.3 ± 1.5 1.3 ± 0.6 1.3 ± 0.6 AUC (µg·h/mL) 1.93 ± 1.27 2.81 ± 0.47 2.14 ± 0.42 3.04 ± 0.13 1.04 ± 0.41

PharmaceuticsPharmaceutics 20212021, 13, 13x FOR, 220 PEER REVIEW 10 of 13 10 of 13

600 HPMCAS ASD (suspension)

500

)

m /

g 400

n Eudragit ASD

(

c n

o 300

m Crystalline ALZ s

a 200

l HPMCAS ASD P

100 PVPVA ASD 0 0 1 2 3 4 5 6 7 Time (hr) FigureFigure 7.7.Plasma Plasma ALZ ALZ sulfoxide sulfoxide concentration concentration profiles afterprofiles oral administrationafter oral administration of crystalline ALZ of crystalline ALZ andand its ASDsASDs in in beagle beagle dogs dogs at a at dose a dose of 10 of mg/kg 10 mg/kg (n = 3). (n Formulations = 3). Formulations were administered were administ as ered as cap- capsules unless otherwise noted as suspension. Symbols: crystalline ALZ (N), PVPVA ASD ( ), sules unless otherwise noted as suspension. Symbols: crystalline ALZ (◆), PVPVA ASD (■), HPM- HPMCAS ASD (N), HPMCAS ASD (suspension) (4), Eudragit ASD (•). CAS ASD (▲), HPMCAS ASD (suspension) (△), Eudragit ASD (●). 4. Discussion Table 4. PK4.1. parameters Dissolution and for LLPS oral Behaviorsadministration of ALZ andstudy Its ASDsof ALZ ASDs in beagle dogs. The LLPS concentration increased with the polymer’s addition, and the highest value Crystalline Eudragit HPMCAS HPMCAS ASD PVPVA Formulations was observed with PVPVA, followed by HPMCAS and Eudragit. If the crystallization tendencyALZ of a drug is low,ASD LLPS is not influencedASD by the presenceSuspension of polymers [8 ]. Thus, ASD this observation indicates that these are values of “apparent” LLPS, which is influenced by Cmax (μg/mL) 0.30 ± 0.21 0.44 ± 0.10 0.34 ± 0.09 0.57 ± 0.05 0.20 ± 0.07 the crystallization of ALZ [7,8]. The occurrence of crystallization was also obvious from Tmax (h) 1.3concentration ± 0.6 profiles1.7 during ± 0.6 the LLPS dissolution2.3 ± 1.5 study (Figure4a). The1.3 “apparent” ± 0.6 LLPS 1.3 ± 0.6 AUC (μg·h/mL) 1.93was ± shown 1.27 to be a good2.81 predictor ± 0.47 of the AUC2.14 of oral± 0.42 absorption controlled3.04 ± by 0.13 solubility [8]. 1.04 ± 0.41 The size of the LLPS particles may also be a factor that influences absorption from ASDs. The particle size measurement based on the light scattering principle is a well-established 4.methodology; Discussion however, the data analysis can frequently be misleading, especially when 4.1.the particlesDissolution have and wide LLPS size distribution.Behaviors of Thus,ALZ weand evaluated Its ASDs the particle size from the dissolution study, where syringe filters with multiple pore sizes were used. It provided directThe information LLPS on concentration the fraction of increasedthe particles with in specific the particlepolymer size’s ranges. addition, Also, and it the highest valueenabled was discussion observed on the with crystallization PVPVA, followed of ALZ during by HPMCAS the LLPS behavior. and Eudragit. If the crystallization ThetendencyµDISS dissolution of a drug study is low, showed LLPS nodissolution is not influenced advantage forby thethe ASDs presence under theof polymers [8]. Thus,acidic pHthis condition, observation whereas indicates the dissolution that these was enhanced are values by formulatingof “apparent” the drug LLPS, with which is influ- encedHPMCAS by orthe PVPVA crystallization under the neutral of ALZ pH [7,8]. condition. The occurrence The observed of concentrations crystallization were was also obvi- well below LLPS concentrations but higher than the equilibrium solubilities. Although the ousreason from for theconcentration effective maintenance profiles of during supersaturation the LLPS in the dissolutionµDISS dissolution study study (Figure rela- 4a). The “apparent”tive to the LLPS LLPS was dissolution shown in to beakers be a isgood unclear, predictor it may have of the been AUC due toof the oral test absorption vessels’ controlled bysmaller solubility scale. As [8]. ALZ The was size not of charged the LLPS under particles the neutral may pH condition,also be a the factor supersaturated that influences absorp- tionstate from was likely ASDs. stabilized The particle by interaction size measurement with polymers, based including on hydrophobicthe light scattering interac- principle is a tion and hydrogen bonding. In the µDISS dissolution study, the ALZ concentration was wellmeasured-established using a UV methodology; probe, which is however, expected to the detect data only analysis dissolved can ALZ. frequently However, in be misleading, especiallythe LLPS dissolution when the study, particles where thehave filtrate’s wide ALZ size concentration distribution. was Thus, determined, we evaluated totally the particle sizedifferent from dissolution the dissolution profiles were study obtained., where The syringe dissolution filters appeared with tomultiple be improved pore with sizes were used. Itthe provided ASDs, and direct Eudragit information appeared to be on the the most fraction effective of polymer. the particles However, in thespecific ALZ particle size ranges.concentration Also, was it enabled affected by discussion the pore size on of thesyringe crystallization filters, which of indicatedALZ during particles’ the LLPS behav- presence in the filtrate. ior. The μDISS dissolution study showed no dissolution advantage for the ASDs under the acidic pH condition, whereas the dissolution was enhanced by formulating the drug with HPMCAS or PVPVA under the neutral pH condition. The observed concentrations were well below LLPS concentrations but higher than the equilibrium solubilities. Alt- hough the reason for the effective maintenance of supersaturation in the μDISS dissolution study relative to the LLPS dissolution in beakers is unclear, it may have been due to the test vessels’ smaller scale. As ALZ was not charged under the neutral pH condition, the supersaturated state was likely stabilized by interaction with polymers, including hy- drophobic interaction and hydrogen bonding. In the μDISS dissolution study, the ALZ concentration was measured using a UV probe, which is expected to detect only dissolved

Pharmaceutics 2021, 13, x FOR PEER REVIEW 11 of 13

ALZ. However, in the LLPS dissolution study, where the filtrate’s ALZ concentration was determined, totally different dissolution profiles were obtained. The dissolution appeared to be improved with the ASDs, and Eudragit appeared to be the most effective polymer. However, the ALZ concentration was affected by the pore size of the syringe filters, which indicated particles’ presence in the filtrate. Moreover, the particles were not likely to be the result of LLPS, but were more likely Pharmaceutics 2021, 13, 220 caused by crystallization. The highest concentration of the Eudragit ASD11 after of 13 filtration with a 0.2 μm filter can be explained by the particles’ wide size distribution, as expected from its large PDI. The LLPS particles’ size was likely to be maintained only in the HPMC solution evenMoreover, after crystallization. the particles were These not findings likely to beare the important result of LLPS, for understanding but were more likely the oral absorption caused by crystallization. The highest concentration of the Eudragit ASD after filtration results.with a 0.2 µm filter can be explained by the particles’ wide size distribution, as expected from its large PDI. The LLPS particles’ size was likely to be maintained only in the HPMC 4.2.solution Relationship even after crystallization.between Dissolution These findings Behavior are and important Oral forAbsorption understanding the oral absorption results. The absorption of ALZ from the ASDs was significantly higher than that from the crystalline4.2. Relationship ALZ between in the Dissolution rat study. Behavior The magnitude and Oral Absorption of effective improvement of the AUC and CmaxThe by absorptionthe polymers of ALZ was from in the the ASDs following was significantly increasing higher order, than that PVPVA, from the HPMCAS, crys- and Eu- dragit,talline ALZ which in the also rat study. agreed The with magnitude that ofof effectivethe LLPS. improvement The relationship of the AUC between and Cmax LLPS concen- trationby the polymers and in wasvivo in AUC the following is presented increasing in Figure order, PVPVA, 8, where HPMCAS, a good andcorrelation Eudragit, was achieved. which also agreed with that of the LLPS. The relationship between LLPS concentration Theand inrelationship vivo AUC is presentedbetween inthe Figure final8, ALZ where concentration a good correlation in wasthe achieved.μDISS dissolution The study in threlationshipe JP2 solution between (pH the final6.8) ALZindicated concentration that the in thesupersaturationµDISS dissolution behavior study in the was JP2 elucidated suc- cessfullysolution (pH. However, 6.8) indicated its that correlation the supersaturation with the behavior AUC was was elucidatedmuch lower successfully. than that with LLPS concentration.However, its correlation The absorption with the AUC of was ALZ much from lower the than ASDs that withwas LLPSlikely concentration. limited by solubility be- The absorption of ALZ from the ASDs was likely limited by solubility because LLPS is causeanalogous LLPS to amorphous is analogous solubility. to amorphous solubility.

60 R² = 0.6582

) 40

L R² = 0.9421

h g

µ 30

(

U A 20

0 0 2 4 6 8 Concentration (µg/mL)

Figure 8. Relationship between final ALZ concentration in the µDiss dissolution study at Figure 8. Relationship between final ALZ concentration in the μDiss dissolution study at pH 6.8 pH 6.8 (open) or LLPS concentration (closed) and AUC in rats. (open) or LLPS concentration (closed) and AUC in rats. Most previously reported oral administration studies of ASDs have been performed with suspensionsMost previously because reported of difficulty oral in administration administering solid studies dosage offorms ASDs for have small been performed withanimals. suspensions However, ASDs because are intended of difficulty to be manufactured in administering as solid solid dosage dosage forms forms when for small ani- they are marketed. Thus, the difference in the performance of suspensions and solid dosage mals.forms hasHowever, been increasingly ASDs are observed intended recently. to be The manufactured most typical approach as solid to fill dosage this gap forms is when they areto improve marketed. the dissolution Thus, the behavior difference of ASDs in using the additivesperformance such as of inorganic suspensions salts [21 ,22and]. solid dosage formsThe has oral been absorption increasingly behavior observed in dogs was recently. completely The different most fromtypical that approach in rats. In to fill this gap isthis to study, improve ASDs werethe dissolution administered behavior as solid dosage of ASDs forms, using which hasadditiv morees practical such sig-as inorganic salts nificance compared to the administration of suspensions. For connecting results from [21,both22]. animal species, the same HPMCAS ASD suspension was evaluated. Despite the large differenceThe oral inabsorption physiologies behavior between rats in anddogs dogs, was a similarcompletely improvement different in the from oral that in rats. In thisabsorption study, was ASDs confirmed were usingadministered the HPMCAS as ASDsolid suspension. dosage forms, The capsule’s which disintegra- has more practical sig- nificancetion process compared appeared to to dramatically the administration affect the oral of absorption,suspensions. as evident For connecting from the large results from both difference in absorption between the two HPMCAS ASDs (suspension and capsule). The animal species, the same HPMCAS ASD suspension was evaluated. Despite the large dif- absorption from the encapsulated solid ASD was comparable to that from the crystalline ferenceALZ. The in Eudragit physiologies ASD allowed between better rats absorption and dogs, than a thesimilar other improvement ASDs, even when in itthe oral absorp- tionwasadministered was confirmed as capsules, using whereas the HPMCAS absorption ASD from the suspension. PVPVA ASD The was worsecapsule than’s disintegration

Pharmaceutics 2021, 13, 220 12 of 13

that from the crystalline ALZ. In this case, the absorption was not likely to have been governed by solubility but rather by the dissolution rate, and LLPS is not a predictor of oral absorption. The absorption profiles of ALZ from capsules were likely to be influenced by particle size of the recrystallized ALZ. That from the Eudragit ASD were permeable to some extent through the 0.2-µm filters (Figure4b), suggesting that small ALZ particles should be available after the recrystallization. As for the HPMCAS ASD, the particle size analysis indicated that the ALZ particles’ mean size after the recrystallization was approximately 0.22 µm (Table2). However, large ALZ crystals might be formed after the dissolution of the PVPVA ASD, which was likely to be the origin of the worse absorption than from crystalline ALZ. This analysis indicates that estimation of the particle size using filters with multiple pore sizes may explain oral absorption if it is limited by dissolution rate. However, for maximizing the efficacy of ASDs, the oral absorption should not be limited by dissolution rate but solubility. In this regard, PVPVA should have the highest potential to improve oral absorption of ALZ based on its highest LLPS concentration. Thus, after optimizing the composition of the binary ASDs, further formulation studies to decide the final dosage form are critical during the developmental study of ASDs. LLPS concentration should work as a parameter that can offer the potential for improvement of oral absorption.

5. Conclusions ALZ was formulated into ASDs using three types of polymers: Eudragit, PVPVA, and HPMCAS, and their LLPS, dissolution, and oral absorption behaviors were investigated. The oral absorption in rats, where the ASDs were administered as suspensions, was not explained by the dissolution test but predicted by the LLPS concentration, which suggested the solubility-limited absorption of ALZ. The efﬁcacy of the ASDs administered to dogs as solid capsules were likely to be limited by dissolution rate. Comprehension of particle size after LLPS appeared to be important for understanding the oral absorption in this case. This observation revealed the challenges in understanding supersaturation, LLPS, and crystallization behaviors occurring in the gastrointestinal tract for interpreting/predicting oral absorption. A formulation strategy that provides sufﬁciently fast disintegration and dissolution for achieving supersaturation was indicated to be important for ASDs.

Supplementary Materials: The following are available online at https://www.mdpi.com/1999 -4923/13/2/220/s1, Figure S1. TG-DTA and Polarized microscopy data for ASDs presented in this study. (a): HPMCAS ASD; (b): PVPVA ASD (3:1); (c): PVPVA ASD (4:1); (d): Eudragit ASD; Table S1. Numerical data for the LLPS dissolution test (Figure4); Table S2. Numerical data for the rat study (Figure6); Table S3. Numerical data for dog study (Figure7). Author Contributions: Conceptualization, K.K. and T.F.; experiments, K.S. and K.K.; discussion, K.S., K.K., M.F., M.O., Y.N., M.M. and T.F.; writing, K.S. and K.K. All authors have read and agreed to the published version of the manuscript. Funding: This research received no external funding. Institutional Review Board Statement: The animal studies were conducted according to the guide- lines of the Declaration of Helsinki and approved by Ethics Committees. Detailed information can be found in “Materials and Methods”. Informed Consent Statement: Not applicable. Data Availability Statement: Detailed data presented in this study are available in supplementary material. Acknowledgments: This study was conducted as part of the Consortium for Biopharmaceutics Tools (CoBiTo) in the Research Center for Drug Discovery and Pharmaceutical Development Science of Ritsumeikan University. Technical supports of the Pharmaceutical and ADMET Research Department, Daiichi Sankyo RD Novare, and KAC Co. are also acknowledged. Conﬂicts of Interest: The authors declare no conﬂict of interest. Pharmaceutics 2021, 13, 220 13 of 13

References 1. Brouwers, J.; Brewster, M.E.; Augustijns, P. Supersaturating drug delivery systems: The answer to solubility-limited oral bioavailability? J. Pharm. Sci. 2009, 98, 2549–2572. [CrossRef] 2. Kawakami, K. Modification of physicochemical characteristics of active pharmaceutical ingredients and application of super- saturatable dosage forms for improving bioavailability of poorly absorbed drugs. Adv. Drug Deliv. Rev. 2012, 64, 480–495. [CrossRef] 3. Paudel, A.; Worku, Z.A.; Meeus, J.; Guns, S.; Van den Mooter, G. Manufacturing of solid dispersions of poorly soluble drugs by spray drying: Formulation and process considerations. Int. J. Pharm. 2013, 453, 253–284. [CrossRef][PubMed] 4. Hens, B.; Kataoka, M.; Ueda, K.; Gao, P.; Tsume, Y.; Augustijns, P.; Kawakami, K.; Yamashita, S. Biopredictive in vitro testing methods to assess intestinal drug absorption from supersaturating dosage forms. J. Drug Deliv. Sci. Technol. 2020, 56, 101275. [CrossRef] 5. Bevernage, J.; Brouwers, J.; Brewster, M.E.; Augustijns, P. Evaluation of gastrointestinal drug supersaturation and precipitation: Strategies and issues. Int. J. Pharm. 2013, 453, 25–35. [CrossRef][PubMed] 6. Bevernage, J.; Brouwers, J.; Annaert, P.; Augustijns, P. Drug precipitation-permeation interplay: Supersaturation in an absorptive environment. Eur. J. Pharm. Biopharm. 2012, 82, 424–428. [CrossRef][PubMed] 7. Kawakami, K.; Suzuki, K.; Fukiage, M.; Matsuda, M.; Nishida, Y.; Oikawa, M.; Fujita, T. Impact of degree of supersaturation on the dissolution and oral absorption behaviors of griseofulvin amorphous solid dispersions. J. Drug Deliv. Sci. Technol. 2020, 56, 101172. [CrossRef] 8. Kawakami, K.; Sato, K.; Fukushima, M.; Miyazaki, A.; Yamamura, Y.; Sakuma, S. Phase separation of supersaturated solution created from amorphous solid dispersions: Relevance to oral absorption. Eur. J. Pharm. Biopharm. 2018, 132, 146–156. [CrossRef] 9. Raina, S.A.; Zhang, G.G.Z.; Alonzo, D.E.; Wu, J.; Zhu, D.; Catron, N.D.; Gao, Y.; Taylor, L.S. Impact of solubilizing additives on supersaturation and membrane transport of drugs. Pharm. Res. 2015, 32, 3350–3364. [CrossRef] 10. Elkhabaz, A.; Moseson, D.E.; Brouwers, J.; Augustijns, P.; Taylor, L.S. Interplay of supersaturation and solubilization: Lack of correlation between concentration-based supersaturation measurements and membrane transport rates in simulated and aspirated human fluids. Mol. Pharm. 2019, 16, 5042. [CrossRef] 11. Kawakami, K. Supersaturation and crystallization: Non-equilibrium dynamics of amorphous solid dispersions for oral drug delivery. Exp. Opin. Drug Deliv. 2017, 14, 735–743. [CrossRef][PubMed] 12. Taylor, L.S.; Zhang, G.G.Z. Physical chemistry of supersaturated solutions and implications for oral absorption. Adv. Drug Deliv. Rev. 2016, 101, 122–142. [CrossRef][PubMed] 13. Indulkar, A.S.; Gao, Y.; Raina, S.A.; Zhang, G.G.Z.; Taylor, L.S. Exploiting the phenomenon of liquid-liquid phase separation for enhanced and sustained membrane transport of a poorly soluble drug. Mol. Pharm. 2016, 13, 2059–2069. [CrossRef] 14. Indulkar, A.S.; Waters, J.E.; Mo, H.; Gao, Y.; Raina, S.A.; Zhang, G.G.Z.; Taylor, L.S. Origin of nanodroplet formation upon dissolution of an amorphous solid dispersion: A mechanistic isotope scrambling study. J. Pharm. Sci. 2017, 106, 1998–2008. [CrossRef] 15. Stewart, A.M.; Grass, M.E.; Brodeur, T.J.; Goodwin, A.K.; Morgen, M.M.; Friesen, D.T.; Vodak, D.T. Impact of drug-rich colloids of itraconazole and hpmcas on membrane flux in vitro and oral bioavailability in rats. Mol. Pharma. 2017, 14, 2437–2449. [CrossRef] 16. Daniel-Mwambete, K.; Torrado, S.; Cuesta-Bandera, C.; Ponde-Gordo, F.; Torrado, J.J. The effect of solubiliation on the oral bioavailability of three benzimidazole carbamate drugs. Int. J. Pharm. 2004, 272, 29–36. [CrossRef][PubMed] 17. Escher, B.I.; Berger, C.; Bramaz, N.; Kwon, J.H.; Richter, M.; Tsinman, O.; Avdeef, A. Membrane-water partitioning, membrane permeability, and baseline toxicity of the parasiticides ivermectin, albendazole, and morantel. Environ. Toxicol. Chem. 2008, 27, 909–918. [CrossRef] 18. Fagerberg, J.H.; Tsinman, O.; Sun, N.; Tsinman, K.; Avdeef, A.; Bergstrom, C.A.S. Dissolution rate and apparent solubility of poorly soluble drugs in biorelevant dissolution media. Mol. Pharm. 2010, 7, 1419–1430. [CrossRef] 19. Kohri, N.; Yamayoshi, Y.; Xin, H.; Iseki, K.; Sato, N.; Todo, S.; Miyazaki, K. Improving the oral bioavailability of albendazole in rabbits by the solid dispersion technique. J. Pharm. Pharmacol. 1999, 51, 159–164. [CrossRef] 20. Castro, S.G.; Sanchez Bruni, S.F.; Urbizu, L.P.; Confalonieri, A.; Ceballos, L.; Lanusse, C.E.; Allemandi, D.A.; Palma, S.D. Enhanced dissolution and systemic availability of albendazole formulated as solid dispersions. Pharm. Dev. Technol. 2013, 18, 434–442. [CrossRef] 21. Keen, J.N.; LaFountaine, J.S.; Hughey, J.R.; Miller, D.A.; McGinity, J.W. Development of itraconazole tablets containing viscous kinetisol solid dispersions: In vitro and in vivo analysis in dogs. AAPS PharmSciTech 2018, 19, 1998–2008. [CrossRef][PubMed] 22. Takano, R.; Maurer, R.; Jacob, L.; Stowasser, F.; Stillhart, C.; Page, S. Formulating Amorphous solid dispersions: Impact of inorganic salts on drug release from tablets containing itraconazole-HPMC extrudate. Mol. Pharm. 2020, 17, 2768–2778. [CrossRef] [PubMed]