DOCTORA L T H E S I S

Department of Health SciencesDivision of Medical Sciences

In Vitro Solubility and Supersaturation Behavior of Supersaturating Drug Delivery

Systems

Amani Alhayali

ISSN 1402-1544ISBN 978-91-7790-154-9 (print)ISBN 978-91-7790-155-6 (pdf)

Luleå University of Technology 2018

Am

ani Alhayali In V

itro Solubility and Supersaturation Behavior of Supersaturating D

rug Delivery System

s

Health Science

In Vitro Solubility and Supersaturation Behavior of Supersaturating Drug Delivery

Systems

A Doctoral Thesis Submitted

By

Amani Alhayali

Division of Medical Sciences Department of Health Sciences Luleå University of Technology

May 2018

Printed by Luleå University of Technology, Graphic Production 2018

ISSN 1402-1544 ISBN 978-91-7790-154-9 (print)ISBN 978-91-7790-155-6 (pdf)

Luleå 2018

www.ltu.se

i

To my family

ii

iii

Abstract The development of new pharmaceutical products has been challenged by the growing number of poorly water-soluble drugs, which often lead to suboptimal bioavailability. Various approaches, such as the use of amor-phous solid dispersions and cocrystals, have been used to improve the solu-bility, and subsequent bioavailability, of these drug molecules. Supersaturat-ing drug delivery systems (SDDSs) have potential for achieving adequate oral drug bioavailability by increasing the drug solubility and creating a su-persaturated state in the gastrointestinal tract. However, there is a need for better understanding of the supersaturation behavior in SDDSs and of the factors affecting supersaturation. The main objective of this thesis was to improve understanding of the supersaturation solubility behavior in SDDSs with a particular focus on rapidly dissolving solid forms (amorphous forms/cocrystals).

In the course of the work, a new formulation for ezetimibe using an amorphous solid dispersion was prepared, cocrystals of tadalafil were pre-pared, and oral films of silodosin were formulated for the first time. These new formulations were thoroughly characterized using a number of solid-state and pharmaceutical characterization techniques.

The dissolution and supersaturation behavior of the prepared SDDSs were studied. The effects of various factors on the supersaturation and precipita-tion characteristics were investigated. These factors included the preparation method, the temperature of the dissolution medium, the type of dissolution biorelevant medium (gastric/intestinal) used, the permeability of the relevant gastrointestinal membranes, the addition of polymers, and the addition of surfactants. The amorphous solid dispersions, cocrystals and oral films that were prepared represent new drug formulations that provide significantly higher dissolution rates and supersaturated solubility than crystalline drug forms. Solid dispersions prepared by the melting method had better super-saturation properties than those prepared by spray drying. The precipitation kinetics of the solid dispersion were faster at 37 ˚C than at 25 ˚C in bio-relevant media. Implementation of an absorption tool during in vitro evalua-tion of supersaturation levels could improve the prediction accuracy of su-persaturation and precipitation. A better understanding of the effects of ex-cipients on the supersaturation and precipitation behavior of these types of formulation was obtained in this thesis. The improvement in supersaturation solubility obtained by adding polymers and surfactants was not proportional to the amounts of excipient used.

iv

This thesis has made notable contributions to the field of pharmaceutical science by advancing our understanding of the supersaturation solubility behavior of the newly prepared SDDSs. Keywords: Poorly soluble drugs; SDDs; Solid dispersions; Cocrystal; Oral films; Super-saturation; Precipitation; Biorelevant media; Dissolution; Absorption.

v

List of Papers This thesis is based on the following papers, which are referred to in the text by their Roman numerals.

I Alhayali A, Tavellin S, Velaga S. (2017) Dissolution and pre-

cipitation behavior of ternary solid dispersions of ezetimibe in biorelevant media. Drug Development and Industrial Pharmacy 2; 43(1):79-88.

II Alhayali A, Selo MA, Ehrhardt C, Velaga S. (2018) Investiga-tion of supersaturation and in vitro permeation of the poorly wa-ter soluble drug ezetimibe. European Journal of Pharmaceuti-cal Sciences, 117, 147-153.

III Shimpi MR, Alhayali A, Cavanagh KL, Rodríguez-Hornedo N, Velaga S. Cocrystals of tadalafil: Physicochemical characteriza-tion, pH-solubility and supersaturation studies. Under revision

IV Alhayali A, Vuddanda PR, Velaga S. Silodosin oral films: De-velopment, physico-mechanical properties and in vitro dissolu-tion studies in simulated saliva. Submitted

Reprints were made with permission from the respective publishers.

Author’s contribution to the included papers

Papers I, II and IV I was responsible for the planning and execution of the work. I prepared the SDDSs and performed the experimental work including the solid-state char-acterization and dissolution studies. I analyzed, plotted, and evaluated the results, and wrote the manuscripts.

Paper III I participated in planning the study and evaluating the results. I performed the experimental work on pH-related solubility and dissolution. I took part in interpreting the results and writing the manuscript. I was not involved in identifying the cocrytals or predicting the theoretical solubility.

vi

vii

Conference Contributions

i Al-Hayali A, Tavelin S, Velaga S. (2014) Dissolution and precipitation behavior of ternary solid dispersions of Ezetimibe in biorelevant media. Poster presentation at: AAPS Annual Meeting and Exposition, San Diego, USA.

ii Alhayali A, Selo MA, Ehrhardt C, Velaga S. (2017) Investi-

gation of supersaturation and permeation of a poorly water soluble drug Ezetimibe. Poster presentation at: 6th FIP Pharmaceutical Sciences World Congress, Stockholm, Swe-den.

viii

ix

Abbreviations A Surface area ANOVA Analysis of variance AP Apical API Active pharmaceutical ingredient BCS Biopharmaceutics Classification System BDDCS Biopharmaceutics Drug Disposition Classification System BL Basolateral C Concentration Ceq Equilibrium solubility concentration Cmax Maximum concentration Cs Saturation solubility Ct Concentration at time t D Diffusion coefficient dM/dt Dissolution rate DMA Dimethyl acetamide DMEM Dulbecco's modified Eagle's medium DMF Dimethyl formamide DMSO Dimethyl sulfoxide DS Degree of supersaturation DSC Differential scanning calorimetry EB Elongation at break EM Elastic modulus EZ Ezetimibe FaSSGF Fasted-state simulated gastric fluid FaSSIF Fasted-state simulated intestinal fluid FBS Fetal bovine serum FDA US Food and Drug Administration FeSSIF Fed-state simulated intestinal fluid FTIR Fourier transformation infrared spectroscopy GIT Gastrointestinal tract h Thickness of the diffusion layer HBSS Hank's balanced salt solution HPC Hydroxypropyl cellulose HPLC High performance liquid chromatography HPMC Hydroxypropyl methylcellulose HPMC-AS Hydroxypropyl methylcellulose acetate succinate HPMCP Hydroxypropyl methylcellulose phthalate ITZ Itraconazole IUPAC International Union of Pure and Applied Chemistry J Absorption/flux

x

Log P Partition coefficient MDSC Modulated differential scanning calorimetry MOA Malonic acid MQ Melt-quenched P Drug permeability coefficient PEG Polyethylene glycol pKa Acid dissociation constant PTFE Polytetrafluoroethylene PVP Polyvinyl pyrrolidone PVP K-30 Polyvinyl pyrrolidone PVPVA Polyvinyl pyrrolidone-vinyl acetate PWSDs Poorly water-soluble drugs PXRD Powder X-ray diffraction SD Spray-dried SDDSs Supersaturating drug delivery systems SDS Sodium dodecyl sulfate SEDDSs Self-emulsifying drug delivery systems SEM Scanning electron microscopy TDF Tadalafil TEER Transepithelial electrical resistance Tg Glass transition temperature TGA Thermo-gravimetric analysis Tm Melting temperature TPGS D-ɑ-tocopheryl polyethylene glycol succinate TS Tensile strength USP United States Pharmacopeia

xi

Contents

1. Introduction ................................................................................................. 1

2. Background ................................................................................................. 3 2.1. Drug bioavailability ............................................................................. 3 2.2. Biopharmaceutical Classification System ........................................... 4 2.3. Drug solubility, supersaturation and dissolution ................................. 5 2.4. Formulation strategies for improving the solubility of poorly water-soluble drugs .................................................................................... 6

2.4.1. Chemical modification................................................................. 7 2.4.2. Physical modification .................................................................. 7

2.5. Supersaturating drug delivery systems (SDDSs) .............................. 14 2.5.1. Spring and parachute theory ...................................................... 14 2.5.2. Precipitation inhibition .............................................................. 15

2.6. In vitro investigation of supersaturation ............................................ 16 2.6.1. Induction of supersaturation in the medium of interest ............. 16 2.6.2. Assessment of drug concentrations in solution as a function of time ...................................................................................................... 17 2.6.3. Factors affecting supersaturation evaluation in vitro ................. 17

3. Aims of the work ...................................................................................... 21

4. Methodology ............................................................................................. 23 4.1. The studied APIs ............................................................................... 23 4.2. Induction of supersaturation by SDDSs ........................................... 23

4.2.1. Preparation of ternary solid dispersion (Paper I) ....................... 23 4.2.2 Preparation of TDF-MOA Cocrystals (Paper III) ....................... 24 4.2.3. Preparation of the casting gel and drug-loaded films (Paper IV) ........................................................................................................ 244.2.4. Solid-state characterization ........................................................ 25

4.3. Induction of supersaturation by the solvent shift method (Paper II) . 27 4.3.1. Supersaturation-precipitation-permeation interplay (without polymer)............................................................................................... 28 4.3.2. Supersaturation-precipitation-permeation interplay (with polymer)............................................................................................... 29

4.4. Preparation of amorphous drug forms (Papers I and III) .................. 29 4.5. Solubility and dissolution studies ...................................................... 29

4.5.1. Solubility study (Papers III and IV) ........................................... 29 4.5.2. Investigation of dissolution and supersaturation solubility (Papers I, III, IV) ................................................................................. 30

xii

4.6. HPLC quantification methods ........................................................... 31 4.7. Statistical analysis ............................................................................. 32 4.8. Preparation of simulated fluids (Papers I, II and IV) ........................ 32

4.8.1 FaSSIF and FaSSGF ................................................................... 32 4.8.2. Simulated saliva ......................................................................... 32

4.9. Cell culture (Paper II) ........................................................................ 32

5. Results and discussion .............................................................................. 35 5.1. Preparation of a ternary solid dispersion of ezetimibe (Paper I) ....... 35

5.1.1. In vitro dissolution study ........................................................... 35 5.2. Preparation of cocrystals of tadalafil (Paper II) ................................ 37

5.2.1. pH-related solubility and Supersaturation index of TDF-MOA cocrystals (Paper III) ........................................................................... 38 5.2.2. Dissolution and supersaturation behavior of TDF solid forms .. 41

5.3. Preparation of silodosin oral films (Paper IV) .................................. 42 5.3.1. In vitro dissolution in simulated saliva ...................................... 44

5.4. Factors affecting supersaturation behavior (Paper I, II, III) .............. 47 5.4.1. Effects of process, formulation and temperature on supersaturation in FaSSIF (Paper I) ..................................................... 47 5.4.2. Effects of process, formulation and temperature on supersaturation in FaSSGF (Paper I) ................................................... 49 5.4.3. Effect of implementation of an absorptive compartment on supersaturation behavior (Paper II) ..................................................... 50 5.4.4. Effect of polymer on supersaturation-precipitation-permeation interplay (Paper II) ............................................................................... 53 5.4.5. Effect of polymer on the supersaturation behavior of cocrystals (Paper III) ........................................................................... 55

6. Conclusions ............................................................................................... 57

7. Future work ............................................................................................... 59

8. Acknowledgments..................................................................................... 61

9. References ................................................................................................. 63

1

1. Introduction Oral drug dosage forms are the most common and the most preferred dosage forms because of the wide range of advantages they provide. Among these are: ease of administration as they can be self-administered by the patient, low production costs compared to other forms, and high patient compliance. Solid oral dosage forms (such as tablets) are associated with greater chemi-cal and physical stability, more accurate doses, and easier handling and manufacturing than other dosage forms (1-4).

Drugs in solid formulations that are administered orally need to be in so-lution (i.e. need to be dissolved) to be absorbed by the gastrointestinal tract (GIT), reach the blood and eventually reach the specific target of action. The extent to which the active pharmaceutical ingredient (API) is available at the site of drug action is defined as the bioavailability (5).

The oral bioavailability of a drug can be affected by several factors, such as the drug’s aqueous solubility, intestinal permeation and metabolism. Poor aqueous solubility is the most frequent cause of low bioavailability because it is necessary for the drug to be dissolved for absorption to take place after oral administration (1, 2). Since the successful therapeutic effect of a drug depends on its concentration in the systemic circulation, higher doses are needed for poorly water-soluble drugs (PWSDs) to achieve the required plasma concentrations. This is often associated with an increased risk of adverse effects (6). The permeability of the intestinal wall to the drug and its absorption through the wall can also affect both the bioavailability and the solubility of the drug in the intestinal lumen (7).

Because the therapeutic activity of the drug is so heavily reliant on its bi-oavailability, it is important to modify some drug properties to improve the therapeutic activity and reduce the incidence and severity of side effects. Various conventional and advanced approaches have been used by formula-tion scientists to improve the solubility, and thereby the bioavailability, of PWSDs (8, 9). Many solubility-enhancement technologies are available, including the use of amorphous solid dispersions, lipids, cocrystals, solubili-zation techniques, surfactants, nanoparticles, cyclodextrins, and others (10, 11). The drug formulations obtained from applying these technologies are called supersaturating drug delivery systems (SDDSs) (10).

Supersaturation is an effective oral bioavailability-improving approach that has been gaining in interest over the last decade as the number of poorly water-soluble compounds has increased in drug-discovery pipelines. SDDSs provide intraluminal drug concentrations that are higher than saturation sol-ubility concentrations (i.e. supersaturation solubility concentrations), thus offering potential for improving bioavailability.

2

The aim of this thesis was to improve the physical properties of some PWSDs. New drug formulations (using SDDSs) with improved physical properties were prepared, to provide drugs with better aqueous solubility and supersaturated drug concentrations (higher than saturation solubility concen-trations) at the GIT.

3

2. Background

2.1. Drug bioavailability Drug bioavailability is the key factor to be addressed when developing a formulation for an oral dosage form. For drugs to be therapeutically active at their site of action they must be absorbed through the intestinal barrier and be sufficiently available in the blood to achieve a therapeutic plasma concen-tration. It is necessary for the drug to be in solution in the gastrointestinal fluid to be well absorbed through the intestinal barrier (Figure 1).

Figure 1. Diagram of the relationship between an oral dose of a drug product and its pharmacological effect.

Thus, factors such as the solubility, absorption and metabolism of a drug can affect its bioavailability (6, 12). An API that is poorly soluble in the GIT and/or is not well absorbed through the intestinal membrane will reach low plasma concentrations after oral administration. As a result, the bioavailabil-ity will be insufficient for the desired pharmacological action. Other factors, such as first-pass metabolism by intestinal cells or liver enzymes (e.g. glucu-ronidation or oxidation by cytochrome P-450 enzymes, sulfation, etc.) can also limit bioavailability (13) (Figure2).

4

Figure 2. Some factors affecting the bioavailability of drugs after oral administration

2.2. Biopharmaceutical Classification System The Biopharmaceutics Classification System (BCS) was developed by Ami-don et al. in 1995 (6). It categorizes pharmaceutical drugs into four groups (Class I, II, III and IV drugs, as shown in Figure 3) depending on their solu-bility and the permeability of the gastrointestinal membrane to the drug. The permeability/flux (J) of a drug through the gastrointestinal membrane is con-trolled by two factors: the drug permeability coefficient (P) of the membrane and the intraluminal gastrointestinal concentration (C) (Equation 1) (10):

J = PC (1)

Solubility is a dose-dependent criterion. For the API to be classified as a soluble drug, the maximum oral dose needs to be soluble in 250 mL of aque-ous medium in the pH range 1 to 7.5 at 37°C, while permeability is consid-ered to be high if 90% of an oral dose is absorbed across the membranes, according to the US Food and Drug Administration (FDA).

The Biopharmaceutics Drug Disposition Classification System (BDDCS) was adapted and modified from the BCS with the focus more on metabolism than on permeability with respect to bioavailability (14, 15). Thus, for PWSDs (BCS classes II and IV), the low intraluminal concentrations of the drug can limit absorption and hence decrease the systemic bioavailability (10).

5

Figure 3. The biopharmaceutical classification system.

2.3. Drug solubility, supersaturation and dissolution The solubility of a solute is defined by the International Union of Pure and Applied Chemistry (IUPAC) as the concentration of the solute in a saturated solution, expressed as the proportion of the designated solute in a designated solvent (16). Solubility occurs in a state of dynamic equilibrium and can be changed by changing the solvent environment (such as temperature or ion content) (17, 18). The term “apparent solubility” describes the solubility at the apparent state of equilibrium between the drug in solution and the drug in an unstable solid state (8, 9). The apparent solubility should not be confused with the equilibrium solubility, which describes the thermodynamic equilib-rium between the drug in solution and the most stable solid form of the drug (10).

When the concentration of drug molecules in the solvent exceeds the equilibrium solubility, the solution is considered to be supersaturated, and this concentration is called the supersaturation solubility. Dissolution is de-fined by the IUPAC as “the mixing of two phases with the formation of one new homogeneous phase” (16). The distinction between solubility and disso-lution is important. The solubility (equilibrium solubility or thermodynamic solubility) is a dynamic property representing the maximum quantity of drug or substance that can be dissolved in a given amount of solvent at a given temperature and pressure. It has nothing to do with the time. The dissolution rate is an indication of the speed of dissolution, a kinetic property.

6

The US Pharmacopeia (USP) lists the solubility terms shown in Table 1. Drugs with solubility lower than 1 mg/ml are described as very slightly sol-uble or practically insoluble. In this thesis, these drugs (with solubility lower than 1mg/ml) are referred to as PWSDs (19).

Table 1. Definitions of solubility according to the US Pharmacopeia (19) Solubility definition Parts of solvent required

for one part of solute Solubility range (mg/ml)

Very soluble <1 >1000Freely soluble From 1 to 10 100-1000Soluble From 10 to 30 33-100Sparingly soluble From 30 to 100 10-33Slightly soluble From 100 to 1000 1-10Very slightly soluble Practically insoluble

From 1000 to 10000 >10000

0.1-1<0.1

2.4. Formulation strategies for improving the solubility of poorly water-soluble drugs Development of new pharmaceutical products for oral administration has been challenged by the increasing number of poorly water soluble molecules in the pipeline. It has been estimated that about 70% of all new APIs emerg-ing from the discovery pipeline have poor aqueous solubility and more than 40% of marketed drugs are poorly water-soluble (20-22). However, most of these drug molecules permeate well through the intestinal wall despite their poor solubility in gastrointestinal fluids (BCS class II) (6, 22). Enhancement of the dissolution rate of BCS II drugs is a key factor in improving their bio-availability because, for the drugs in this group, the only limitations to im-proved bioavailability are the dissolution rate and solubility, and slight in-creases can sometimes make noticeable improvements (2).

The solubility of class II drugs can be improved by various formulation strategies that induce intraluminal concentrations higher than saturation sol-ubility concentrations (supersaturation). Commonly used strategies include solid dispersions, self-emulsifying drug delivery systems (SEDDSs), cy-clodextrin complexes, cocrystals and nanocrystals. These strategies are based on either chemical or physical modification of the drug molecule. Ta-ble 2 provides a list of the most common formulation strategies for improv-ing the solubility of PWSDs; this thesis discusses two formulation strategies: amorphous solid dispersions and cocrystals.

7

Table 2. Formulation strategies for improving the solubility of poorly water-soluble drugs (BCS II)

Chemical modification • Prodrugs• Salts

Physical modification • Cocrystals• Particle size reduction• Solid dispersion• Porous materials• Carriers (cyclodextrines, liposomes)

2.4.1. Chemical modification 2.4.1.1. Prodrugs The use of prodrugs is a well-known method for improving the solubility of PWSDs. The molecular structure of the API is chemically modified to form an inactive derivative that is swallowed, metabolized by enzymes, and trans-formed into the parent pharmacologically active form of the drug in the GIT (23).

2.4.1.2. Salts Most drugs are either weak acids or weak bases and the dissolution rates and solubility of these drugs in aqueous media are often improved if the drugs are in the form of a salt. The disadvantages of using the salt form include relatively poor stability and increased incidence of undesirable side effects which may affect patient compliance (24, 25).

2.4.2. Physical modification The Nernst-Brunner equation (Equation 2), a modified version of the Noyes-Whitney equation, is the cornerstone for improving the solubility and disso-lution rates of BCS II drugs (26).

𝑑𝑀𝑑𝑡

= 𝐴𝐷(𝐶𝑠−𝐶𝑡)ℎ

(2)

This equation states that the dissolution rate (dM/dt) is directly propor-tional to the saturation solubility (Cs), the concentration of the drug in the medium at any time (Ct), the available surface area of the dissolving solid (A) and the diffusion coefficient of the compound (D). The thickness of thediffusion layer is expressed as h. Each parameter in the equation is affectedby many factors and manipulation of these factors will modify the dissolu-tion rate of the drug, as shown in Figure 4.

8

The equation indicates that the dissolution rate can be improved by manipu-lation of the saturation solubility and the surface area. Most of the physical methods that modify the physical properties of APIs and increase their solu-bility are based on the Nernst-Brunner equation. These methods include increasing the surface area of the dissolved drug by reducing the drug parti-cle size, increasing the apparent solubility of the drug, and improving the wetting and solubilizing properties of the compound to reduce the thickness of the boundary layer around the drug particles (27).

2.4.2.1. Particle-size reduction One of the oldest methods for improving drug solubility involves reducing the particle size. According to the Noyes-Whitney and Nernst-Brunner equa-tions, a decrease in particle size results in a larger surface area of drug avail-able for solvation and a subsequent increase in the dissolution rate and po-tential bioavailability of PWSDs (28). Particle size reduction to < 1 μm has been associated with a significant increase in solubility (28, 29). Several methods have been reported for reducing drug particle size to generate nano-sized particles; these include milling, spray drying and homogenization (30, 31). The particle size reduction method is considered to be safe as the API is used without any additives or chemical modifications, which implies that there will not be any new adverse effects.

Figure 4. Physicochemical factors affecting the dissolution rate according to the Nernst-Brunner equation.

9

2.4.2.2. Amorphous state

Solids can be either amorphous or crystalline (4). In the crystalline state the molecules are structured regularly in a lattice formation. Compounds with an irregular molecular arrangement that lacks long range order are referred to as being in the amorphous state. Van der Waals forces, hydrogen bonding and other intermolecular interactions hold the molecules within the crystal lat-tice, resulting in lower molecular mobility. Because the molecular interac-tions in the crystalline state are strong, the compound is usually more stable but the dissolution rate is slower.

When a crystalline drug is heated, it undergoes melting at temperature Tm; then when the molten drug is slowly cooled, the molecules have sufficient time to move from their current location to a thermodynamically stable point on the crystal lattice, regenerating a crystalline structure (32). If, however, the molten drug is cooled suddenly, it can reach a super-cooled liquid state. If it is cooled further below its Tm, the system remains in equilibrium until the glass transition temperature (Tg) is reached, below which it enters a non-equilibrium (rubbery) state and converts into the glassy (amorphous) state of the drug. The glass transition is a second order thermodynamic transition characterized by a step change in the heat capacity which is also associated with changes in other thermodynamic properties such as volume, enthalpy, and entropy (33). The amorphous state of a drug has a higher free energy than the crystalline state and molecules or atoms can convert gradually into the highly ordered crystalline state if they are maintained at a specific tem-perature for a long time.

Amorphous solids have higher dissolution rates than the crystal forms and, when they are dissolved, can result in higher drug concentrations in solution than the saturated solution of the crystalline form (i.e. they can form a supersaturated solution). This phenomenon has been used to increase the solubility of various PWSDs (34-36). The drug in supersaturated solution will eventually regain its equilibrium solubility, as the amorphous form will convert to the stable crystalline form. Therefore, maintaining the PWSD in its amorphous state is one of the great challenges in formulation develop-ment.

2.4.2.3. Amorphous solid dispersions Pure amorphous drugs are rarely developed alone as pharmaceutical prod-ucts because their high levels of free energy make them thermodynamically unstable. Therefore, finding ways to stabilize the amorphous form is seen as important. Polymer and surfactant excipients, included to prepare and stabi-lize the amorphous form, have created tremendous opportunities for the pharmaceutical scientist to address issues relating to the bioavailability of poorly soluble molecules. Solid dispersions in general can be defined as

10

molecular mixtures of PWSDs in hydrophilic carriers (20). Solid dispersions are classified into three types (first, second or third generation) according to their development and the carrier/excipient used in the formulation.

First generation solid dispersions are prepared using crystalline carriers. Urea and sugars were the first crystalline carriers employed in solid disper-sions. The associated improvements in solubility were attributed to faster carrier dissolution, releasing microcrystals or particles of drug. These solid dispersions have the disadvantage that the drug is not released as quickly as from the amorphous form, despite being more stable thermodynamically (37, 38).

The second generation of solid dispersions contains amorphous carriers instead of crystalline carriers. The drugs are molecularly dispersed in an irregular form within an amorphous carrier, usually consisting of a polymer, to create the most common type of solid dispersion used to date, amorphous solid dispersions (39). In second generation solid dispersions, the drug is in its supersaturated state because of forced solubilization in the carrier at a molecular level. The dissolution of the polymer dictates the drug release profile (27, 40).

Third generation solid dispersions contain a surfactant carrier or a mixture of amorphous polymers and surfactants as carriers with the aim of stabilizing the solid dispersion and achieving the highest degree of bioavailability for PWSDs. Surfactants have surface activity or self-emulsifying properties, and thus help to improve the dissolution properties of PWSDs and potentially improve their bioavailability. Surfactants such as inulin, inutec SP1, com-pritol 888 ATO, gelucire 44/14 and poloxamer 407 have been used success-fully as carriers to increase the bioavailability of various PWSDs (41-45). A combination of surfactants and polymers has also been used to improve dis-solution and prevent precipitation of some PWSDs (46, 47).

A solid dispersion, therefore, can now be defined as a dispersion of amor-phous drug in a polymeric carrier matrix. Selection of a suitable carrier is critical to obtaining a solid dispersion with improved dissolution properties. Drug solubility in the carrier and drug-carrier compatibility are the main factors to be considered when selecting a carrier. Phase separation and drug precipitation can result from inadequate solubility of the drug in the carrier. Drug-carrier incompatibility can cause formulation failure (9, 27, 48). Melt-ing and solvent evaporation methods are mainly used to prepare solid disper-sions. Processes that have been used in manufacturing solid dispersions are shown in Figure 5 (20).

Polymers

The inclusion of polymers in the formulation is one of the most common methods of stabilizing an amorphous API (49). Polymers appear to be the most successful carriers for solid dispersions because of their ability to form amorphous solid dispersions.

11

Figure 5. The manufacturing processes used in the preparation of solid dispersions (20).

Polymers improve the physical stability of the amorphous compounds by increasing the Tg, which delays the crystallization process. Polymer carriers not only stabilize the dispersed amorphous drug but can also increase wetta-bility and prevent drug precipitation in the GIT, as found recently (50). Pol-ymers can be divided into two types: (1) natural product-based polymers which are mainly composed of cellulose derivatives such as hydroxypropyl methylcellulose (HPMC), hydroxypropyl cellulose (HPC), hydroxypropyl methylcellulose acetate succinate (HPMC-AS), hydroxypropyl methyl cellu-lose phthalate (HPMCP), ethylcellulose, or starch derivates such as cy-clodextrins (40, 51-56); and (2) synthetic polymers such as polyvinyl pyrrol-idone (PVP), polyethylene glycols (PEG) and polymethacrylates (39, 57-59).

Surfactants

The chemical structures of surfactant (surface active agent) molecules have two distinct ends: one end is hydrophilic and the other is hydrophobic. Be-cause of this amphiphilic property, surfactants can be adsorbed onto the sur-face or interface of a system between two phases, e.g. a water-oil interface. Addition of surfactants to amorphous drug formulations has the potential to reduce nucleation (crystal growth) and crystallization as a result of either alteration of the viscosity of the molecule or changes to the interfacial ener-gy at the amorphous solid interface. However, the ability for surfactants to increase the solubility of PWSDs or to stabilize amorphous APIs in formula-tion depends on factors such as the chemical structure of the drug and the surfactant, the temperature, and the pH (60, 61).

12

2.4.2.4. Oral films Oral films have received increasing interest in the pharmaceutical industry, especially in recent years. They provide a wide range of potential ad-vantages, including improved compliance, suitability for pediatric and geri-atric use, and avoidance of first-pass metabolic effects (62, 63): Compliance: Oral films are easy to administer without chewing or intake of water as the films disintegrate and dissolve rapidly to release the drug when placed in the oral cavity. This is thought to have potential to improve patient compliance. Pediatric and geriatric use: Oral films are useful for special patient groups such as pediatric and geriatric patients because the potential for choking is lessened in patients who have difficulties in swallowing solid oral dosage forms like tablets and capsules. Fast-dissolving oral films provide a solution for these patients and for patients with mental disorders such as mania and schizophrenia, and patients with dysphagia or emesis (63). Metabolism problems: Some drugs are metabolized or degraded in the gas-trointestinal environment. This undesirable effect can be avoided by formu-lating drugs as oral films which will be absorbed directly into the systemic circulation without passing through the GIT, thus avoiding first-pass me-tabolism in the liver (64, 65).

Oral dissolving films are composed of an API dispersed within film-forming polymers and plasticizers. They also contain excipients for sweeten-ing, flavoring, coloring and saliva stimulation. Polymers may be used as drug release platforms. Both natural and synthetic polymers (such as Pullu-lan, starch, HPMC, polyvinyl alcohol and PVP) can be used in the formula-tion of oral films in different ratios to obtain effective drug delivery plat-forms. Modification of the composition of the polymeric matrix can help to achieve desirable formulation properties such as faster disintegration and better mechanical properties (66, 67).

Plasticizers can improve the flexibility of the film and reduce brittleness. They reduce the Tg of the polymers and enhance the film-forming properties (68). Commonly used plasticizers include PEG, glycerol, diethyl phthalate, triethyl citrate, tributyl citrate, etc (69, 70).

Surfactants such as Tweens, sodium lauryl sulphates and Polaxamer 407 play an important role as solubilizing, wetting, and dispersing agents that enable films to disintegrate within seconds, releasing the incorporated API immediately (70, 71).

Sweeteners and flavoring agents could be significant in improving palata-bility and compliance for pediatric and geriatric patients. Sucrose, dextrose and glucose have been used as sweeteners. Excipients such as peppermint oil, cinnamon oil, oil of nutmeg, white vanilla, chocolate, and citrus also provide flavor (72).

13

The increase in saliva from the addition of saliva-stimulating agents such as ascorbic acid, citric acid and malic acid can promote the disintegration of oral films (70, 71).

Sublingual, fast-dissolving films are oral films that are intended to be placed under the tongue in the mouth. The relatively high vascularity and permeability of the sublingual mucosa and membrane can result in rapid absorption of the formulated drug and instant bioavailability, resulting in a rapid onset of action (73).

Fast-disintegrating film formulations can be used to improve the bioavail-ability of the drugs and to overcome the solubility limitations in aqueous media. Improvements in the solubility and dissolution of the drug are the result of the drug being converted from crystalline to amorphous form and dispersed in the hydrophilic polymer and also as a result of the large exposed surface area of the film. Solvent casting and extrusion methods are the most frequently used methods of preparing oral films.

2.4.2.5. Cocrystals Cocrystals are multicomponent solids that contain two or more components in a single homogeneous crystalline system with well-defined stoichiometry. A cocrystal is composed of an API and a coformer. The APIs can be weakly acidic, basic or even neutral compounds while the coformer is a benign non-toxic molecule (such as malonic acid, saccharin, nicotinamide, etc.) or an-other API. Cocrystal formation has often been guided by hydrogen-bonded assemblies between the neutral molecules of the drug and the cocrystal coformer (74, 75).

Over recent decades, many pharmaceutical cocrystals have been prepared and these have received significant interest from the pharmaceutical indus-try. The most significant application of pharmaceutical cocrystals is im-provement in the solubility of PWSDs. The highly soluble coformer, dis-solves to a greater extent than the drug, creating supersaturation with respect to cocrystal as a result of nonstoichiometric concentration in solution (75-78).

The drug delivery systems obtained as a result of applying these technol-ogies and performing physical modifications to alter the properties of the PWSD are called SDDSs. An amorphous solid dispersion of ezetimibe (EZ; a PWSD) was prepared during the course of the thesis, with the aim of im-proving its solubility and dissolution. Cocrystals of tadalafil (TDF; a PWSD) were also prepared to improve the dissolution and solubility properties of the powder. In addition, a sublingual silodosin film was prepared to deliver the drug oromucosally. These films also have potential to provide supersaturated concentrations of the drug in the saliva.

14

2.5. Supersaturating drug delivery systems (SDDSs) SDDSs are promising tool for improving the oral bioavailability of PWSDs. SDDSs provide drugs with intraluminal concentrations that are higher than the saturation solubility concentrations (i.e. supersaturation). The rationale behind SDDSs is based on Fick’s First Law (Equation 1), i.e. the enhanced concentrations at the site of absorption can increase the flux of the formulat-ed drugs across the gastrointestinal membrane. SDDSs are thermodynami-cally stable or metastable in the formulation. They generate a thermodynam-ically metastable, supersaturated solution of the drug after dissolution in the aqueous environment of the GIT.

The drug is in the supersaturated state when its solubility is higher than its equilibrium solubility in the given medium. The high concentration is achieved by incorporating the drug in a high energy state such as an amor-phous phase in amorphous solid dispersions (53) or highly soluble forms such as cocrystals (79) in the SDDS.

The drug molecules in the supersaturated state tend to precipitate rapidly as this is a thermodynamically unstable state. Thus, the value of SDDSs in improving solubility and absorption will depend on the extent and duration of maintaining the supersaturated state in the gastrointestinal lumen. There-fore, the challenge is in creating concentrations of the drug that are many times higher than the thermodynamic solubility concentration, and maintain-ing these high concentrations for an adequate time period for the absorption process to take place.

The stabilization and inhibition of precipitation are accomplished by us-ing precipitation inhibitors such as polymers, surfactants, and other excipi-ents that act as stabilizers of supersaturated drugs in the SDDS (80). The generation and maintenance of the metastable supersaturated state are repre-sented by the ‘spring and parachute’ theory (81) in SDDSs.

2.5.1. Spring and parachute theory The benefit of supersaturation as a strategy for improving the intestinal ab-sorption and bioavailability of PWSDs depends on two essential steps: the generation (the ‘spring’) and maintenance (the ‘parachute’) of the metastable supersaturated state, as described by Guzman et al. (81) and shown in Figure 6. The ‘spring’ is formed by using a higher energy form of the PWSD (com-pared to the crystalline form), which can be achieved using many formula-tion options (such as the amorphous form or cocrystals, as described previ-ously). Once supersaturation has been induced, precipitation of drug mole-cules will occur through kinetically or thermodynamically controlled pro-cesses.

The ‘parachute’ is the temporary inhibition of precipitation by interfering with nucleation and/or crystal growth, achieved by the use of pharmaceutical

15

excipients or other components called precipitation inhibitors (10, 82). Pre-cipitation inhibitors are commonly used to maintain supersaturation and inhibit drug precipitation for an extended period of time (83, 84).

Figure 6. Schematic illustration of the concentration-time profiles showing the ‘spring and parachute’ theory of supersaturated drug delivery systems (10).

2.5.2. Precipitation inhibition Identification of the optimum precipitation inhibitor is a vital aspect of suc-cessful SDDS formulations. Polymers such as cellulose derivatives (e.g. HPC, HPMC), vinyl polymers [e.g. PVP, PVP-vinyl acetate (VA)], and eth-ylene polymers (e.g. PEG) have been used to stabilize supersaturation (85-89). Precipitation inhibitors used for stabilizing a supersaturated solution act by a variety of mechanisms depending on the properties of the inhibitor, the drug and the medium. The mechanisms usually involve increasing the solubility and/or decreasing nucleation and crystal growth, by increasing the viscosity (which results in reduced molecular mobility, thus decreasing nucleation and crystal growth), adsorption onto the crystal surface (which hinders crystal growth), or changing the level of solvation at the crystal/liquid interface (which slows the incorporation of drug molecules into a crystal lattice) (10).

However, some polymers can increase the solubility of drugs (90-93). Surfactants can also delay or hinder precipitation from supersaturated solu-tions when added at concentrations exceeding their critical micelle concen-tration. The increase in drug solubility reduces the rate of nucleation and crystal growth. Examples of surfactants are sodium dodecyl sulfate (SDS), D-ɑ-tocopheryl polyethylene glycol succinate (TPGS), and Poloxamers (10).

16

2.6. In vitro investigation of supersaturation There are many assays for evaluating supersaturation and precipitation or precipitation inhibition. They differ in their approach to the generation of supersaturation, the techniques used for supersaturation measurement, and the experimental conditions. Evaluation of supersaturation in vitro requires the induction of supersaturation in the medium of interest and then assess-ment of drug concentrations in solution as a function of time.

2.6.1. Induction of supersaturation in the medium of interest Supersaturation can be induced and evaluated for either formulated (i.e. SDDSs) or non-formulated drugs.

2.6.1.1. Formulated drugs (SDDSs) In a SDDS, supersaturation is induced as an inherent characteristic of the formulation and there is no need for it to be induced as part of the assay. Evaluation of the supersaturation behavior of a SDDS requires an aqueous medium that is relevant for the environment in the GIT (94). One compart-ment/one phase setups traditionally used to study dissolution are based on USP I or II apparatus. One-compartment and two-compartment pH shift approaches have been used to evaluate the dissolution of formulations that rely on the gastrointestinal pH gradient to induce supersaturation, simulating gastric and intestinal dissolution behavior (95-97). A multi-compartment dissolution setup that includes gastric, intestinal and absorption compart-ments has also been used to predict supersaturation and precipitation in SDDSs (98).

2.6.1.2. Non-formulated drugs Evaluation of the supersaturation-precipitation potential of non-formulated drugs (APIs) or the precipitation inhibition capacity of an excipient requires induction of supersaturation as a starting step. There are several methods of measuring supersaturated drug concentrations; the most commonly applied are the solvent-shift and pH-shift methods (90).

In the solvent-shift method, the PWSD is first dissolved in a solvent that is water miscible and has a significantly higher solubilizing capacity for the drug than the aqueous medium in which supersaturation is to be evaluated [e.g. dimethyl sulfoxide (DMSO), dimethyl formamide (DMF) and dimethyl acetamide (DMA)] (99-101). Next, a fraction of this solution is added to the medium under investigation. Finally, supersaturation and/or precipitation can be evaluated. This is a common and simple method of creating supersatura-tion and is applicable to any PWSD that can be dissolved at significantly higher concentrations in a water miscible solvent (102, 103).

17

The pH-shift method is used to evaluate the supersaturation potential of ion-izable drugs. The solubility of drugs can increase in polar aqueous solvents as a result of ionization; therefore, any shift in the pH that reduces ionization will decrease the drug solubility and induce a supersaturated state. This method can be considered more biorelevant for weakly basic drugs as a pH shift occurring upon transfer of drugs from the stomach to the small intestine can induce supersaturation in vivo (104, 89).

2.6.2. Assessment of drug concentrations in solution as a function of time Assessment of supersaturated concentrations resembles classic concentration assessments during dissolution testing. The measured drug concentrations are combined with the equilibrium solubility of the drug in the test medium (which includes the precipitation inhibitor if one has been added to the drug formulation). Accordingly, the drug concentration can be measured and ex-pressed relative to the equilibrium solubility as the degree of supersaturation (DS), as shown in Equation 3: DSt = Ct/Ceq (3)

Where Ct is the drug concentration at time t and Ceq is the equilibrium solubility of the drug in the test medium. The saturation extent can be quantified as a measure of the thermodynamic tendency for precipitation (90). DS < 1: subsaturated DS = 1: saturated DS > 1: supersaturated

2.6.3. Factors affecting supersaturation evaluation in vitro 2.6.3.1. Formulation-related factors The dissolution characteristics of SDDSs in vitro are mainly affected by the extent of supersaturation maintenance, which is in turn determined by 1) the preparation method and 2) the formulation composition (105). Methods used for preparing solid dispersions, such as quenching of hot melts, spray drying, and film casting, could play a role in stabilizing formulated amorphous drugs in SDDSs.

In hot melt quenching, mixtures of the drug and excipient carrier are heat-ed to a molten state, followed by immediate cooling and solidification. This is an effective method of transforming a drug to a critical supersaturated state while avoiding nucleation (50, 106). The supersaturation of a solid dispersion containing itraconazole (ITZ) and low-grade HPMC-AS powder

18

(HPMC-AS-LF) by the melting method had better results than using amor-phous ITZ, which was physically unstable and recrystallized at high temper-ature and humidity levels. No physical or chemical instability was observed in the ITZ and HPMC-AS-LF solid dispersion, which was attributed to the temperature- and moisture-activated electrostatic interactions between the drug and their counter ionic polymers (107).

Spray-dried dispersions had the single Tg of a drug/polymer mixture ac-cording to solid-state characterization. The dissolution rate and supersatura-tion maintenance were increased, which was attributed to the forced aggre-gation of the polymer into regular micelle structures (108). Some other methods, such as the amorphization strategies, that were used to prepare solid dispersions also had positive results (109).

Formulation compositions such as the polymer type and the presence of specific drug:polymer ratios are important for maintaining the stability of the supersaturated state. When the polymer content within the solid dispersion is increased, the DS will be increased significantly (110, 111). According to investigations of the solid dispersions prepared from three polymers (HPMC, PVP, and PEG 6000), formulations incorporating HPMC maintained super-saturation for the longest period (112). The polymer HPMC-AS as precipita-tion inhibitor had the best effect of 41 types of polymer and surfactant in another study (113). Synergistic enhancement of amorphous stability and dissolution of the drug can be obtained by combining two polymers in ter-nary solid dispersions (114) or a polymer and a surfactant (115).

2.6.3.2. Assay-related factors The experimental conditions that are chosen for in vitro assays during super-saturation evaluation could affect the outcome of the assay significantly and thus require careful consideration. These conditions include the following:

Medium selection

Biorelevant dissolution media that simulate fasted and fed conditions in the stomach and small intestine have been developed (116). These provide more accurate predictions of dissolution than simple aqueous buffer solutions. The accuracy of in vitro/in vivo correlations has been improved significantly with the use of biorelevant media (117).

The components present in gastrointestinal fluids, including bile salts and phospholipids, could affect the precipitation kinetics. Bevernage et al. have compared the precipitation behavior of PWSDs in simulated and human gastrointestinal fluids. They explored the importance of careful selection of dissolution medium for supersaturation assays and found that simple aque-ous buffer solutions should be avoided when studying precipitation kinetics in the intestinal environment as they overestimate the stability of supersatu-ration. Fasted-state simulated intestinal fluid (FaSSIF) is a reasonable medi-

19

um for predicting precipitation behavior in fasted-state human fluids. In con-trast, fed-state simulated intestinal fluid (FeSSIF) can significantly underes-timate precipitation. Fasted-state simulated gastric fluid (FaSSGF) should be used in place of USP simulated gastric fluid to predict precipitation behavior in the gastric environment as the latter underestimates precipitation (99, 118, 119).

Temperature

Most of the in vitro supersaturation assays are performed at 25 ˚C or room temperature instead of the more biologically relevant temperature of 37 ˚C. Alonzo et al. investigated the dissolution behavior of amorphous felodipine at 25 ˚C versus 37 ˚C. Amorphous felodipine induced supersaturation during dissolution at 25 ˚C but not at 37 ˚C (120). The altered dissolution behavior of felodipine resulted from temperature-dependent drug crystallization kinet-ics after contact with the dissolution medium at different temperatures.

Implementing an acceptor/absorption compartment

Addition of an acceptor compartment to simulate absorption during evalua-tion of SDDSs is crucial and will improve the predictive power of the assay. Permeation and absorption can affect the dissolution and precipitation kinet-ics of the PWSD (121). Permeation into a sink compartment may relieve the thermodynamically unstable supersaturated system and act as an alternative for precipitation. As demonstrated by Bevernage et al., precipitation of loviride was significantly reduced in the presence of an absorption compart-ment in the Caco-2 model when compared to a one-compartment setup with-out an absorption compartment (122).

Different experimental approaches have been used to simulate absorption, including the addition of an immiscible organic layer as an absorptive sink, or the integration of an actual “absorption” compartment separated from the dissolution medium by a filter/pump combination or a Caco-2 monolayer (121, 123).

Hydrodynamics

The hydrodynamics of the supersaturation assays will influence the nuclea-tion and crystal growth processes of drug molecules (124, 125). In general, extensive mixing and increased kinetic energy can assist in overcoming the activation obstacle for nuclei formation (10, 126). Very few studies have addressed this aspect of hydrodynamics and its effect on supersaturation and precipitation behavior with respect to PWSDs. Carlert et al. compared the in vitro precipitation of a PWSD (AZD0865) in a stirring model versus a shak-ing model (95). The precipitation rate was significantly slower in the shaking

20

model, indicating the importance of hydrodynamics in supersatura-tion/precipitation evaluation assays. The mixing that is usually applied during in vitro supersaturation assays is expected to be relatively high compared to in vivo gastrointestinal motility in the fasted state (117, 121). Therefore, it has been hypothesized that in vitro assays will often overestimate precipitation (127), and more research is needed to investigate the nature of in vivo hydrodynamics and its implemen-tation in supersaturation/precipitation models (128). Understanding and ap-plying in vivo hydrodynamics could significantly improve the biorelevance of in vitro supersaturation and precipitation evaluation studies.

21

3. Aims of the work The main aim of this thesis was to improve understanding of the supersatura-tion solubility behavior of SDDSs in vitro. Another aim was to study the effects of various experimental factors on supersaturation and precipitation behavior in vitro. These aims were achieved by formulating and evaluating SDDSs. This was accomplished by the following specific aims:

• To prepare SDDSs (amorphous solid dispersions, cocrystals, oral films) of the PWSDs EZ, TDF and silodosin and to characterize their physical and mechanical properties (Papers I, III, IV)

• To investigate the solubility and supersaturation behavior of the

formulated products in buffers, FaSSIF and FaSSGF, and simulated saliva (Papers I-IV)

• To induce supersaturation by two strategies, i.e. using formulated

and non-formulated drug systems (Papers I-IV)

• To investigate the effects of various factors, such as preparation

method, medium type, medium temperature (25ºC or 37ºC), and permeation across Caco-2 cell monolayers, on the supersaturation and precipitation behavior of PWSDs (Papers I-II)

• To investigate the effects of excipients (polymers and surfactants) on

the supersaturation and precipitation behavior of SDDSs. The poly-mers studied were PVP K-30, HPMC and HPMC-AS. The surfac-tants used were Poloxamer 188 and TPGS (Papers I-IV).

22

23

4. MethodologyA more detailed description of the methodology can be found in the “Mate-rials and Methods” sections of the individual publications.

4.1. The studied APIs The drugs used in papers I, II, III and IV were EZ (Sigma-Aldrich, Stock-holm, Sweden)), TDF (Gift from Yonsei University, South Korea) and silo-dosin (Ultra Medica, Damascus, Syria). The criteria for selection of drug compounds to be investigated in Papers I-IV were that the drugs: a) should be poorly water soluble, mainly BCS class II compounds; b) should be stable under the selected experimental conditions, especially at the pH used for the solubility and permeability determinations; c) should be used in their con-firmed crystalline form and e) should be transported mainly by passive means if included in the permeability study through the Caco-2 cell mono-layers.

4.2. Induction of supersaturation by SDDSs 4.2.1. Preparation of ternary solid dispersion (Paper I) 4.2.1.1. Melt-quenching method Prior to the preparation of solid dispersions, EZ and PVP K-30 were kept in the oven at 100 ˚C for 24 h to remove any adsorbed water from the solids, and differential scanning calorimetry (DSC) analysis was performed to en-sure the crystalline nature of the dried EZ. Crystalline EZ, PVP K-30 and poloxamer 188 in various proportions (see Table 3) were accurately weighed and mixed gently for 1–2 min using a porcelain mortar and pestle. The re-sulting powder mixtures were placed in a preheated oven at 200 ˚C for 5–8 min. After complete melting, the liquid was immediately put into a freezer at ˗20˚C for 1–2 days. The melt-quenched (MQ) samples were then stored in a desiccator over silica until the day of analysis. Before the analysis, all the formulations were ground and sieved using 125 mm sieves.

24

4.2.1.2. Spray-drying method Crystalline EZ, PVP K-30 and poloxamer 188 in various proportions (see Table 3) were accurately weighed and dissolved (equivalent to 2% w/v of solids) in methanol. The solutions were spray dried using a Buchi mini spray dryer B-29 attached to an inert loop B-295 to trap the residual solvents. The processing conditions were as follows: inlet temperature 80 ˚C, air flow 357 L h˗1, aspiration 70–80%, feed-flow rate 3 mL/min and outlet temperature 40–50 ˚C. Nitrogen gas was used as the drying gas in the closed-loop exper-iment. The collected spray-dried powders were stored in a sealed desiccator over silica until the day of analysis.

Table 3. Proportions of the components in the melt-quenched (MQ) and spray-dried (SD) solid dispersions Sample Ezetimibe (%w/w) PVP K-30 (%w/w) Poloxamer188 (%w/w)

MQ1 or SD1 5 90 5 MQ2 or SD2 15 80 5 MQ3 or SD3 10 80 10

4.2.2 Preparation of TDF-MOA cocrystals (Paper III) A 7-mL glass vial was charged with 389 mg (1 mmol) of TDF and 104 mg (1 mmol) of malonic acid (MOA). Ethyl acetate 3ml was added to form slur-ry. The reaction was allowed to stir for a total of 5 hours. The formation of a solid cake indicated product formation. After 5h, the solids were isolated by vacuum filtration and air dried at room temperature to yield 419 mg (85%) of TDF-MOA cocrystals.

4.2.3. Preparation of the casting gel and drug-loaded films (Paper IV) The casting gel consisted of HPMC or HPMC-AS (65%), glycerol and pro-pylene glycol (7%), and sweetener (2%), with ethanol and water as vehicle, relative to the total weight of the solid base. All weights are w/w ratios. Silodosin was dissolved in ethanol (12-13%) and the remaining excipients were dissolved in water (87-88%). HPMC or HPMC-AS was gradually add-ed to this solution under constant magnetic stirring (800 rpm) at ambient temperature (21± 1 ˚C) until a homogeneous gel was obtained. This casting gel was kept for 6–12 h to remove the air bubbles. Table 4 shows the overall composition of the prepared films.

10g of the gel was cast onto a fluoropolymer-coated polyester sheet §using an automated film applicator equipped with a coating knife (Coat-master 510, Erichsen, Sweden). The silodosin dose of 8 mg was loaded intoeach 6 cm2 film by fixing the wet film thickness at 750μm with a casting

25

speed of 5 mm/s. The cast films were dried in a convective hot-air oven (Binder, Sweden) at 60 °C for 45-50 min. After drying, the films were care-fully peeled off, sealed in plastic (polythene) zip pouches, and stored in a desiccator (23 °C/40% RH) until further characterization.

Table 4. Formulations of silodosin oral films Formulation code Drug: polymer ratio

F1 1:5 (Silodosin:HPMC) F2 1:3 (Silodosin:HPMC) F3* 1:5 (Silodosin:HPMC) F4* 1:3 (Silodosin:HPMC) F5 1:5 (Silodosin:HPMC-AS) F6 1:3 (Silodosin:HPMC-AS) F7* 1:5 (Silodosin:HPMC-AS) F8* 1:3 (Silodosin:HPMC-AS) *Film formulated using TPGS (0.5 % w/w)

4.2.4. Solid-state characterization 4.2.4.1 Differential scanning calorimetry (DSC) and modulated differential scanning calorimetry (MDSC) The thermal behavior of the raw materials and SDDSs (Papers I, III and IV) was studied using a DSC Q1000 (TA Instruments, New Castle, DE). The instrument was calibrated before use for temperature and enthalpy using indium. The sample was accurately weighed and placed in a non-hermetic aluminum pan, which was then crimp-sealed. The samples were heated from 25 ˚C to 200 ˚C (EZ), 25 to 320 ˚C (TDF) and 25 to 120 ˚C (silodosin) at a heating rate of 10˚C/min under continuous nitrogen purge (50 mL/min). The DSC Q1000 was equipped with a refrigerated cooling system. The data were analyzed using TA analysis software (TA Instruments, New Castle, DE).

MDSC was carried out (Paper I) using a Q1000 instrument (TA Instru-ments, New Castle, DE) for better identification of the Tg through the separa-tion of reversible (i.e. Tg) from nonreversible (i.e. enthalpy recovery, fusion) thermal events. Samples (2–5 mg) were accurately weighed and placed in aluminum non-hermetic pans, which were then crimped. These samples were then heated from 25 ˚C to 200 ˚C at 5˚C/min with modulation amplitude of ±0.80 ˚C every 60 s.

4.2.4.2. Powder X-ray diffraction (PXRD) PXRD patterns were obtained (Papers I, III and IV) using an Empyrean PXRD instrument (PANalytical, Almelo, The Netherlands) equipped with a Pixel3D detector and a monochromatic Cu-Kα radiation X-ray tube (1.54056 Å). The tube voltage and amperage were 45 kV and 40 mA, respectively.

26

Powdered samples (Papers I and III) were loaded into the oval cavity in the metal sample holder and carefully leveled. The experimental settings were as follows: 2θ ranged from 5˚ to 40˚, step size 0.02˚ 2θ. The data were pro-cessed using High Score plus Version 3.0 software (PANalytical, Almelo, The Netherlands).

The instrument was calibrated using a silicon reference standard. For Pa-per IV, film samples (3 × 3 cm2 films) were placed on a silicone (zero back-ground) plate which was fitted into the sample metal holder. Samples were scanned at a diffraction angle of 2θ between 5˚and 40˚, increasing in step sizes of 0.02º. All patterns were obtained at 25 ± 1 °C.

4.2.4.3. Thermo-gravimetric analysis (TGA) The moisture content and thermal degradation behavior (Papers I, II and IV) of all the formulation components were investigated using a Thermal Ad-vantage TGAQ5000 (TA Instruments, New Castle, DE) connected to a cool-ing system. Samples were placed in the platinum pans and the furnace was heated at a rate of 5˚C/min. Nitrogen at a flow rate of 50 mL/min was used as a purging gas (balance gas and sample gas: 10 and 90 ˚C min-1, respective-ly). Universal analysis software was used to analyze the results (TA Instru-ments, New Castle, DE).

4.2.4.4. Scanning electron microscopy (SEM) The morphology (Papers I and IV) of the samples was examined using a Merlin scanning electron microscope (Zeiss, Oberkochen, Germany) equipped with X-Max 50 mm2 X-ray detectors (Oxford Instruments, Abing-don, UK). All the samples were coated with tungsten to increase the conduc-tivity of the electron beam. The instrument voltage was 15 kV and the cur-rent was 1 nA.

4.2.4.5. Fourier transformation infrared spectroscopy (FTIR) A Bruker IFS66v/S FTIR spectrometer equipped with a deuterated triglycine sulfate detector was used to obtain the FTIR spectra of the samples in Paper I. Powdered samples were mixed with KBr and IR spectra were obtained in the diffuse reflectant mode (Kubelka Munk). The following experimental settings were used: 64 scans with a resolution of (4 cm-1), spectral region 400-4000 cm-1.

4.2.4.6. Raman spectroscopy The Raman spectra for cocrystals prepared in Paper III were recorded on a Chromex Sentinel dispersive Raman unit equipped with a 785nm, 70 mW excitation laser and a TE cooled CCD. Each spectrum is a result of 20 co-added 20-second scans. The unit had continuous automatic calibration using

27

an internal standard. The data were collected by Sentinel Soft data acquisi-tion software and processed in GRAMS AI.

4.2.4.7. Oral film characterization Mechanical properties The dynamic mechanical strength of the films was tested using a hybrid rhe-ometer in DMA mode (DHR2, TA Instruments, Sweden). Briefly, samples of cast films were cut into rectangular strips of 1×5 cm2 and 1 cm at each end was held between clamps; thus, the effective testing area was 1×3 cm2. The upper clamp was then used to stretch the film upwards at a constant linear rate of 0.1 mm/min until the film ruptured. Stress and strain were computed by Trios® software.

The tensile strength (TS; Equation 4) and the elongation at break (EB; Equation 5) were obtained from the peak stress and the maximum strain, respectively, in the stress vs strain plot. Tensile tests are commonly used to determine the robustness of film preparations. The TS is the maximum force applied to the film sample at the breaking point and the EB is the length of the film during the pulling process. In addition, Young´s modulus or the elastic modulus (EM) describes the influence of the strain and its force at this strain on the film area. The EM was obtained from the initial elastic deformation region in the stress vs strain plot (129).

𝑇𝑒𝑛𝑠𝑖𝑙𝑒 𝑠𝑡𝑟𝑒𝑛𝑔𝑡ℎ (𝑇𝑆) = 𝑃𝑒𝑎𝑘 𝑠𝑡𝑟𝑒𝑠𝑠𝐶𝑟𝑜𝑠𝑠−𝑠𝑒𝑐𝑡𝑖𝑜𝑛𝑎𝑙 𝑎𝑟𝑒𝑎 𝑜𝑓 𝑡ℎ𝑒 𝑓𝑖𝑙𝑚

(4)

𝐸𝑙𝑜𝑛𝑔𝑎𝑡𝑖𝑜𝑛 𝑎𝑡 𝑏𝑟𝑒𝑎𝑘 (𝐸𝐵) = 𝐼𝑛𝑐𝑟𝑒𝑎𝑠𝑒 𝑖𝑛 𝑙𝑒𝑛𝑔𝑡ℎ 𝑎𝑡 𝑏𝑟𝑒𝑎𝑘𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑓𝑖𝑙𝑚 𝑙𝑒𝑛𝑔𝑡ℎ

× 100 (5)

Disintegration time Samples (1 × 1 cm2) were placed in a Petri dish containing 2 mL of water and shaken at 60 rpm using an orbital shaker water bath at 37 ± 1 °C. The disintegration time of the films was evaluated using a modified Petri dish method (130). The time to disintegration or disruption was measured with a stopwatch.

4.3. Induction of supersaturation by the solvent shift method (Paper II) Specific amounts of EZ were dissolved in DMSO and four stock drug solu-tions were prepared. The solvent shift method was used to induce different degrees of supersaturation (DS10, DS20, DS30 and DS40), based on the equilibrium concentration of EZ in FaSSIF. For example, the concentration of the solution at DS10 was 10 times the equilibrium solubility concentration of EZ.

28

4.3.1. Supersaturation-precipitation-permeation interplay (without polymer) 4.3.1.1. One-compartment setup A one-compartment experimental setup (without Caco-2 cells) was used to investigate the supersaturation and precipitation behavior of EZ in the ab-sence of a Caco-2 cell monolayer. The supersaturation experiments were conducted in 12-well plates (22.1mm diameter). FaSSIF (0.5 mL) was added to each well and supersaturation was induced using the solvent shift method by adding specific volumes of drug/DMSO stock solution. The final concen-tration of DMSO in FaSSIF was <1%. The other experimental conditions were as follows: 60 rpm shaking speed and 37 °C temperature. Samples were withdrawn at fixed time points and centrifuged immediately, and the ob-tained supernatants were diluted and analyzed using high performance liquid chromatography (HPLC).

4.3.1.2. Two-compartment setup Apical compartment (donor compartment) FaSSIF was used as the supersaturation medium in the apical (AP) com-partment. The solvent shift method was used to induce DS10, DS20, DS30 and DS40.

Basolateral compartment (acceptor compartment). Hank’s balanced salt solution (HBSS) was the basis of the transport medium in the basolateral (BL) compartment of the cell monolayer. The HBSS was buffered with HEPES buffer solution 10 mM to a pH of 7.4 and supplement-ed with 25 mM glucose. TPGS was added to the transport medium in a con-centration of 0.2% to provide sink conditions.

4.3.1.3. Vectorial transport experiments The transport of EZ across Caco-2 cell monolayers in Transwell inserts was investigated. Samples from the AP side (40 μL) were taken at 0, 5, 15, 30, 45, 60, 90 and 120 min and centrifuged immediately at 17,000 g for 5–10 min. The supernatants were diluted with the mobile phase and analyzed us-ing HPLC to find the concentration of EZ. The amount of EZ in the AP compartment (0.5 mL) was quantified. Samples (200 μL) were withdrawn from the BL side at the same time points and analyzed without dilution. Fresh buffer (200 μL) was added to the BL side at each time point to ensure the maintenance of sink conditions. Each plate set of Transwell membrane inserts was kept in an agitating incubator with an orbital shaker (speed 60 rpm and medium temperature 37 ˚C) and was only taken out for sampling.

29

4.3.2. Supersaturation-precipitation-permeation interplay (with polymer) The effect of the polymer on supersaturation and permeation was studied in one-compartment and two-compartment setups with the DS40 sample. PVP K-30 was pre-dissolved in FaSSIF in two concentrations: 0.05 and 0.1% w/v on the AP side of the two-compartment setup and in the one-compartment setup. The DS40 supersaturated solution of EZ was prepared based on the equilibrium solubility of the drug in FaSSIF in the presence of the polymer.

4.4. Preparation of amorphous drug forms (Papers I and III) Crystalline EZ (Paper I) was accurately weighed and placed in a preheated oven. After complete melting, the liquid was immediately put into a freezer at ˗20 ˚C. The material was then gently ground with a mortar and pestle un-der controlled water vapor pressure and immediately analyzed by DSC and PXRD. Before the analysis, all the formulations were ground and sieved using 125 mm sieves. Amorphous TDF (Paper III) was prepared by spray drying a 2% w/v solution of TDF in acetone:water (9:1, v/v), as outlined by Wlodarski et al. (131). A Buchi mini spray dryer B-290 (Büchi, Switzerland) was used. An open sys-tem was used with nitrogen as the drying and atomizing gas (open, suction mode). The spray drying conditions were as follows: inlet temperature 160 ˚C, pump speed 30 %, and outlet temperature 89 ˚C. The aspirator was oper-ated at 100%. The obtained powder was sieved (500 micron) and analyzed by DSC and PXRD to confirm the amorphous state. The powders were stored at -20 ˚C until further analysis.

4.5. Solubility and dissolution studies 4.5.1. Solubility study (Papers III and IV) 4.5.1.1. Cocrystal solubility and determination of eutectic concentrations of TDF and MOA In paper III, the eutectic point between cocrystal and drug was reached by simply adding 200 mg of cocrystal powder (500 microns) to 5 mL of solu-tion at different pHs. A 0.1 N HCL solution was used to obtain pHs of 1, 2 and 5, an acetate buffer solution was used for pHs 3 and 4, and a phosphate buffer solution was used for pHs 6 and 7. The solutions were subjected to orbital shaking (60 rpm) for 48-72 hours at 25˚C and the pH at equilibrium was measured. The concentrations of the drug and the coformer at equilibri-um at the eutectic point were determined by HPLC. The eutectic point was confirmed by PXRD analysis of the residual solid sample at equilibrium.

30

The equilibrium solubility of TDF was measured at a range of pHs from 1 to 7; the pH of the solution was adjusted by 1 M HCl or NaOH. An excess amount (approx. 200 mg) of crystalline TDF was added to a 5 mL solution of 0.1 N HCl and slurried in a water bath at 25 ˚C with orbital shaking (60 rpm.) After equilibration for 48-72 hours, the sample solutions were filtered [using 0.2 µm polytetrafluoroethylene (PTFE) filters] and analyzed by HPLC (Agilent Technologies, Stockholm). The pH at equilibrium was measured and the excess solids were analyzed using PXRD to confirm the solid form at equilibrium.

4.5.1.2. Equilibrium solubility determination of silodosin The solubility of silodosin was determined in simulated saliva containing pre-dissolved HPMC or HPMC-AS with or without TPGS in concentrations similar to those used in the film formulations. An excess of drug was added to conical flasks containing 10 mL of saliva and the other polymer-surfactant excipients. The flasks were tightly closed and placed in a shaker water bath at 37°C. After 48 h, the separated aliquots were filtered through a 0.45 μm filter, diluted appropriately and analyzed using HPLC. Each experiment was performed in triplicate (n = 3) and the results were reported as means ± standard deviation.

4.5.2. Investigation of dissolution and supersaturation solubility (Papers I, III, IV) 4.5.2.1. Dissolution of ternary solid dispersions in simulated intestinal fluids Excess amounts of the solid dispersions of EZ (Paper I) were suspended in 10mL of FaSSIF or FaSSGF in sealed test tubes and placed in a preheated shaking water bath at an orbital shaking speed of 60 rpm. The experiments were performed at 25 ˚C and 37 ˚C. The samples were withdrawn after 5, 15, 30, 45, 60, 75, 90, 120, 180 and 240 min and centrifuged (5 min, 17,000×g), filtered through 0.2 mm PTFE syringe filters to ensure complete separation of the solid phase, and analyzed by HPLC. For comparison, the dissolution properties of crystalline and amorphous EZ in FaSSIF and FaSSGF were also studied. The samples were withdrawn at pre-defined intervals following the above procedure.

4.5.2.2. Dissolution of cocrystals in phosphate buffer Non-sink dissolution experiments were conducted to investigate the dissolu-tion and supersaturation behavior of amorphous TDF and TDF-MOA co-crystals (Paper III). Excess amounts (approx. 1 mg/mL) of crystalline TDF, amorphous TDF and an equivalent amount of the cocrystal powder were suspended in a phosphate buffer solution (pH 6.8). The experiments were

31

conducted at 25 ˚C with 60 rpm orbital shaking. Aliquot samples were with-drawn at specific time points up to 3 hours, filtered (0.2 µm PTFE filters), diluted and analyzed by HPLC.

The dissolution of crystalline TDF, amorphous TDF and the cocrystals was also investigated in the presence of HPMC. HPMC was pre-dissolved in the phosphate buffer solution (0.1% w/v; pH 6.8) and the dissolution exper-iments were conducted under similar conditions to those above. At the end of the dissolution study, the pH of the solutions was measured and the sepa-rated solids were analyzed with PXRD.

4.5.2.3. Dissolution of oral films in simulated saliva Non-sink dissolution studies were carried out in simulated saliva at pH 6.8. The films (samples F1-F8; 2 × 3 cm) were carefully dropped into 10 mL dissolution medium under continuous orbital shaking (60 rpm) at 37˚C. Samples of the medium were then withdrawn at different times, filtered through syringe filters (0.45 μm) and analyzed by HPLC. The DS was calcu-lated from the drug concentrations at different times during dissolution of the films (F1-F8) and the drug concentrations at equilibrium. DS calculations for the formulated drug were based on Equation 3 (DS=Ct/Ceq). The obtained values of DS for F1-F8 were plotted versus time.

4.6. HPLC quantification methods Solutions were analyzed by reversed phase HPLC (HPLC 1290 series in-strument equipped with a quaternary pump and a UV detector; Agilent, San-ta Clara, CA). An isocratic method was used.

For determination of the concentration of EZ (Papers I and II), a C-18 Eclipse plus 3 mm (4.6mm ×100mm) column was used. The mobile phase consisted of 40:60 (v/v) 0.05% sodium 1-heptane sulfonic acid and acetoni-trile with a flow rate of 0.5 mL/min. The injection volume was 10 μL and the measurements were carried out at a wavelength of 230 nm.

TDF and MOA were quantified using a C-18 Nucleosil column (150×4.6 mm, 5 microns; Sigma Aldrich, Sweden) for both compounds. For TDF, the mobile phase was 20 mM Na-phosphate buffer (pH 7) and acetonitrile (30:70). Analysis was performed at a flow rate of 0.5 mL/min and a detector wavelength of 260 nm. For MOA, the mobile phase was 25 mM potassium dihydrogen phosphate (pH 2.5) and methanol in a ratio of 93:7. The flow rate and wavelength were 1 mL/min and 210 nm, respectively.

Sample separation for silodosin (Paper IV) was performed on an Agilent Eclipse-plus C18 column (5 μm, 250 mm × 4.6 mm) with a mobile phase of 25 mM potassium dihydrogen phosphate buffer (pH 7.0) and acetonitrile 40:60 (v/v) at 25˚C. The flow rate was 1 mL/min and the determination wavelength was 269 nm.

32

4.7. Statistical analysis One-way analysis of variance (ANOVA) with Tukey's post hoc multiple comparisons and the t-test were used to determine statistically significant differences (p < 0.05) in Papers II and IV. All results were expressed as means with standard deviation (n=3). Mechanical properties in Paper IV were tested with n = 5.

4.8. Preparation of simulated fluids (Papers I, II and IV) 4.8.1 FaSSIF and FaSSGF FaSSIF/FeSSIF/FaSSGF powder from biorelevant.com (London, UK) was used to prepare the simulated intestinal fluid. FaSSIF (sodium taurocholate, lecithin, sodium chloride, sodium hydroxide and monobasic sodium phos-phate) and FaSSGF (sodium taurocholate, lecithin, sodium chloride and hy-drochloric acid) were prepared and the pH adjusted to 6.5 (FaSSIF) and 1.5 (FaSSGF) with NaOH and HCL.

4.8.2. Simulated saliva Simulated saliva was prepared in the laboratory by dissolving the ingredients in purified water (Table 5) and adjusting the pH to 6.8 with NaOH (132).

Table 5. Simulated saliva formulation (132)

Ingredients Grams/liter of purified water

Sodium chloride (NaCl) 0.126 Potassium chloride (KCl) 0.964 Potassium phosphate monobasic (KH2PO4) 0.655 Urea 0.200 Sodium sulfate (Na2SO4 10H2O) 0.763 Ammonium chloride (NH4Cl) 0.178 Calcium chloride dihydrate (CaCl2 2H2O) 0.228 Sodium bicarbonate (NaHCO3) 0.631

4.9. Cell culture (Paper II) Caco-2 cells were routinely cultured in 75 cm2 tissue culture flasks (Greiner Bio One, Frickenhausen, Germany) at 37 °C with 5% CO2 atmosphere in high glucose Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 20% fetal bovine serum (FBS), 1mM sodium pyruvate, 100 U/mL peni-cillin and 100 μg/mL streptomycin. For the transport studies, the cells were

33

seeded at a density of 75,000 cells/cm2 on rat-tail collagen type I-coated Transwell clear inserts (12mm in diameter, pore size 0.4 μm; Corning, VWR, Dublin, Ireland). The cells were allowed to grow for at least 16 days and the medium was exchanged every other day.

The transepithelial electrical resistance (TEER) of the cell monolayers was measured using a Millicell ERS-2 epithelial volt-ohm meter equipped with STX-2 electrodes (Millipore, Carrigtwohill, Ireland) and corrected for the background value contributed by the cell culture inserts and medium. To exclude any effects of the sampling process on the cells' integrity and to ensure the tightness of the inter-cellular junctions, the TEER values for all cell monolayers were measured before starting and at the end of each exper-iment.

34

35

5. Results and discussion

5.1. Preparation of a ternary solid dispersion of ezetimibe (Paper I) The cholesterol-lowering drug EZ was chosen as the model drug in this study. EZ is a lipophilic molecule with a partition coefficient (logP; oc-tanol/water) of 4.5. It is very weakly acidic (close to neutral) with an acid dissociation constant (pKa) of 10.2 (aromatic hydroxyl group) and thus is practically insoluble in water. EZ has been classified as a class II drug; its low dissolution rate in the gastric lumen could potentially lead to dissolu-tion-limited bioavailability (133-135).

In this study, ternary solid dispersions of EZ were prepared using differ-ent proportions of the polymer PVP K-30 and the surfactant poloxamer 188. Two preparation methods were employed: melt quenching and spray drying. Visually glassy materials of pure EZ and the solid dispersions were obtained using the melt-quenching method. The formulations containing EZ, polox-amer and polymer obtained with the spray-drying technique were dry, white powders of EZ and the solid dispersions.

Solid-state characterization of the solid dispersions of EZ was carried out using DSC, MDSC, PXRD and FTIR. The amorphous nature of the solid dispersions was confirmed in all the characterization tests. A single Tg (no melting peak for EZ) was observed at 140–180 ˚C for all the solid disper-sions when examined by MDSC, suggesting that EZ was molecularly dis-persed within the polymer and surfactant to form an amorphous solid solu-tion (Figure 7) (35, 136, 137).

SEM images of the solid dispersions showed that the two preparation methods resulted in different particle morphologies for the same formula-tion, while there were no major differences in particle morphology between the different formulations prepared by the same method. Further, there were no EZ crystals on the surfaces of the particles (Figure 8).

5.1.1. In vitro dissolution study (Paper I) It is important to optimize the in vitro experimental conditions in studies

of the supersaturation behavior of amorphous solids to ensure that they are biorelevant to the in vivo conditions. This study used biorelevant test media, non-sink conditions and orbital shaking. It has been suggested that non-sink conditions are more biorelevant for evaluating the dissolution behavior of supersaturable formulations where the maximum solution concentrations are determined (138)

36

Figure 7. Modulated differential scanning calorimetry reversing heat-flow thermograms of (a) melt-quenched (MQ) solid dispersion 1, (b) MQ2 and (c) MQ3; and (d) spray-dried (SD) solid dispersion 1, (e) SD2 and (f) SD3 showing the heat capacity (Tg) in the region 140-180 °C.

The effect of hydrodynamics on supersaturation/precipitation is poorly understood. In a recent paper, the precipitation rate was shown to be remark-ably slower in the shaking model than in the stirring model, emphasizing the importance of the applied hydrodynamic methodology (95). The factors af-fecting supersaturation are discussed in more detail later in the thesis.

37

Figure 8. Scanning electron microscope photographs of (a) crystalline ezetimibe (EZ), (b) amorphous EZ, and (c) melt-quenched (MQ1) and (d) spray-dried (SD1) solid dispersions of EZ.

5.2. Preparation of cocrystals of tadalafil (Paper III) The drug TDF was used as the study drug for preparing cocrystals in this study, with MOA as the coformer. TDF is a phosphodiesterase inhibitor, a pharmaceutically active ingredient marketed under the brand name Cialis and approved by the FDA in 2003 for the treatment of erectile dysfunction. TDF is a white crystalline powder, practically insoluble in water that has been classified as a class II drug by the BCS. It is a neutral drug, with a pKa of 15.17. The TDF logP (octanol:water) is 1.54, which reflects poor drug solubility in water (139-142).

The pharmaceutical cocrystals of TDF were formed with the coformer MOA, as confirmed by solid-state characterization using Raman spectrosco-py, PXRD and DSC. All these techniques unequivocally suggested the for-mation of TDF-MOA cocrystals. For example, the diffraction pattern for the new form was distinctly different from that of TDF and MOA separately, as shown in Figure 9.

38

Figure 9. Powder X-Ray Diffraction patterns for (a) tadalafil (TDF), (b) TDF-MOA cocrystals and (c) malonic acid (MOA). The asterisks denote new diffraction peaks in the cocrystal form in comparison with the individual cocrystal components.

5.2.1. pH-related solubility and Supersaturation index of TDF-MOA cocrystals (Paper III) Understanding the pH dependence of cocrystal solubility helps to guide the development of formulations for cocrystals as well as helping to determine their supersaturation index and conversion/transformation propensity. TDF-MOA (1:1) cocrystals consist of non-ionic drug, TDF and one diprotic acid MOA, with two pKa values (2.83 and 5.69). The TDF-MOA cocrystal solu-bility dependence on [H+] is given by Equation 6:

𝑆𝑇𝐷𝐹−𝑀𝑂𝐴 = �𝐾𝑠𝑝 �1 + 𝐾𝑎1,𝑀𝑂𝐴[𝐻+] + 𝐾𝑎1,𝑀𝑂𝐴𝐾𝑎2,𝑀𝑂𝐴

[𝐻+]2 � (6)

Where 𝐾sp is the cocrystal solubility product, and 𝐾a1 and 𝐾a2 are the ionization constants for MOA. 𝐾sp was calculated as the product of the con-centrations of the non-ionized constituents according to Equation 7:

(7)

Where [𝑇𝐷𝐹]𝑇 and [𝑀𝑂𝐴]𝑇 are the eutectic concentrations of TDF and MOA, respectively. The TDF-MOA cocrystals converted back into TDF at pHs in the range of 1 to 7.4. The solubility was measured at the eutectic point where the solution phase was in equilibrium with both cocrystals and solid drug phases (75). The cocrystal 𝐾sp was determined from Equation 7.

𝐾𝑠𝑝 =[𝑇𝐷𝐹]𝑇[𝑀𝑂𝐴]𝑇

�1 +𝐾𝑎1,𝑀𝑂𝐴

[𝐻+] +𝐾𝑎1,𝑀𝑂𝐴 × 𝐾𝑎2,𝑀𝑂𝐴

[𝐻+]2 �

39

Table 6 summarizes the cocrystal 𝐾sp and intrinsic solubility of the drug, STDF.

𝐾sp was used to generate the dependence of the cocrystal solubility on pH. Figure 10a shows that the solubility of the TDF-MOA cocrystals was higher than that of TDF over a wide pH range. For example, even at their lowest solubility, the cocrystals appeared to be over 100 times more soluble than the drug (Figure 10a and Table 6). These trends are consistent with the coformer solubility and ionization properties, in line with the results reported for other cocrystals (75, 143).

Table 6. Summary of the tadalafil (TDF)-malonic acid (MOA) cocrystal solubility product (Ksp ± standard deviation), the intrinsic solubility of TDF (STDF), and the cocrystal:TDF solubility ratio (SCC/STDF). The solubility was measured under non-ionizing conditions

Cocrystal 𝑲𝐬𝐩 (M2) STDF (M) SCC/STDF (pH 1-3) TDF-MOA cocrystal (3.8 ± 1.4 ) × 10-07 (4.6 ± 0.2) × 10-6 134-210

The experimental cocrystal solubility could only be measured in the pH range 1 to 3, as the coformer controlled/lowered the pH of the solution. The mathematical model based on 𝐾sp and 𝐾a values, (Equation 6) predicted an exponential increase in cocrystal solubility with pH, and these theoretical values were in agreement with the experimentally measured solubilities (Figure 10a). Thus, it is possible to predict the dependence of the cocrystal solubility on pH from the Ksp, at pH values where the solubility was not experimentally measured, and to gain insight into the supersaturation behav-ior of the cocrystals at physiologically relevant pHs. It is clear from the shape of the solubility-pH curve that the formation of TDF-MOA cocrystals changed the solubility of TDF to being pH-dependent, demonstrating the potential of cocrystal technology in engineering the solubility behavior of non-ionizing drugs as a function of pH.

Cocrystals of saccharin have previously been shown not only to enhance the aqueous solubility of saccharin but also to impart pH sensitivity to co-crystallized non-ionizing drugs like carbamazepine (143) Furthermore, the TDF eutectic concentrations were in agreement with the drug solubility vaues (STDF; Figure 10a), suggesting that there was no drug complexation with the coformer in solution.

The enhanced solubility of cocrystals has been previously reported for a number of drugs (144, 145). The solubility advantage or supersaturation index (SA), is defined as Scc/Sdrug, where Scc and Sdrug are cocrystal and drug solubility, respectively. The SA represents the driving force for drug precipi-tation and provides a measure of the risk of cocrystal conversions. It also provides guidance for the selection of pH and formulation additives to pre-vent drug precipitation. Supersaturation index points were calculated from

40

experimental cocrystal solubility and eutectic drug concentrations. The theo-retical SA was extrapolated according to Equation 8:

𝑆𝑇𝐷𝐹−𝑀𝑂𝐴

𝑆𝑇𝐷𝐹=

�𝐾𝑠𝑝�1+𝐾𝑎1,𝑀𝑂𝐴�𝐻+�

+𝐾𝑎1,𝑀𝑂𝐴𝐾𝑎2,𝑀𝑂𝐴

�𝐻+�2 �

𝑆𝑇𝐷𝐹=

�𝐾𝑠𝑝�1+𝐾𝑎1,𝑀𝑂𝐴�𝐻+�

+𝐾𝑎1,𝑀𝑂𝐴𝐾𝑎2,𝑀𝑂𝐴

�𝐻+�2 �

[𝑇𝐷𝐹]𝑇(8)

Figure 10b shows the dependence of the SA on pH for TDF-MOA co-crystals. The theoretical and experimentally measured values (in the range pH 1-3) are in good agreement. Interestingly, the SA for TDF-MOA cocrytals increased from approximately 100 at pH 1 (measured) to about 48,000 at pH 6.8 (predicted) (Figure 10b).

Figure 10. (a) pH dependence of the solubility of 1:1 tadalafil-malonic acid (TDF-MOA) cocrystals and TDF. Experimentally measured solubilities of the cocrystals and drug are represented by red circles (○) and blue squares (□), respectively. [TDF]T concentrations are shown as blue circles (○). The lines represent the theo-retical solubilities of the cocrystals (red dashed line) and drug (blue line). The de-pendence of the solubility on pH was calculated according to Equation 6 using the 𝐾𝑠𝑝 value listed in Table 6 and pKa values of 2.83 and 5.69 for MOA. The solubility of TDF (STDF ; line) was determined as the average of the experimental TDF solubil-ity points (blue squares). (b) The cocrystal supersaturation index (cocrystal solubili-ty advantage) as a function of pH for TDF-MOA cocrystals. The symbols represent the calculated cocrystal:TDF solubility ratios from the solubility measurements and the line represents the theoretical dependence on pH (Table 6).

The SA values are several orders of magnitude higher than the experi-mentally measured supersaturation generated by amorphous or cocrystal forms because of the onset of nucleation at a lower supersaturation value. In studying the SA of cocrystals, Good and Rodriguez-Hornedo have shown that the solubility of cocrystals is directly proportional to the solubility of the constituent reactants for a wide range of drugs (including acidic, basic, zwit-

41

terionic and neutral drugs) (75). Further, Velaga et al. revealed an increase in the coformer concentration at the eutectic point for carbamazepine-saccharin cocrystals and concluded that the aqueous SA of the cocrystals was the result of a decrease in the solvation energy (log γ) (143).

5.2.2. Dissolution and supersaturation behavior of TDF solid forms (Paper III) Figure 11 shows non-sink dissolution profiles for TDF cocrystals, TDF amorphous and crystalline solids in the presence and absence of the precipi-tation inhibitor HPMC at pH 6.8. In the absence of HPMC, TDF in the form of cocrystals and the amorphous form dissolved faster and reached supersat-uration sooner, followed by precipitation.

The solubility of the crystalline form of TDF reached a plateau corre-sponding to the equilibrium solubility of the drug (0.09× 10-05 M) in phos-phate buffer at pH 6.8. The spray dried TDF (amorphous form) and cocrys-tals dissolved very quickly and reached maximum concentrations (Cmax) of ∼ 0.9 × 10-05 M and 2.61 × 10-05 M, respectively, in less than 5 min. There-after, the concentrations decreased because of precipitation or conversion ofthe cocrystals to TDF, as confirmed by PXRD analysis. The dissolution be-havior of TDF-MOA cocrystals resembles that of many other cocrystals(144). The maximum concentration of a drug can be expressed in terms of itssupersaturation (Cmax/Sdrug). Cmax/Sdrug thus represents the concentrationabove which precipitation occurs faster than the dissolution of the cocrystals.The cocrystal and amorphous forms generated a supersaturation level of ∼30and 10, respectively (Figure 11).

42

Figure 11. Dissolution profiles of different solid forms of tadalafil (TDF) in phos-phate buffer (pH 6.8±0.4): crystalline TDF (blue), amorphous TDF (green), TDF-malonic acid (MOA) cocrystals (red); and in phosphate buffer containing 0.1% w/v HPMC: amorphous TDF (black) and TDF-MOA cocrystals (purple). Error bars represent standard deviations obtained from triplicate measurements.

5.3. Preparation of silodosin oral films (Paper IV) Silodosin is a selective α1A adrenoceptor blocker that is a safe, effective treatment for the relief of voiding and storage symptoms in patients with benign prostatic hyperplasia (146). It is a white to pale yellowish powder with a logP (octanol/water) of 2.87, pKa1 of 8.53 and pKa2 of 4.03. The bioavailability of oral silodosin capsules is nearly 32% (147). The FDA label states that silodosin is very slightly soluble in water. Oral films of silodosin were successfully prepared with HPMC or HPMC-AS as the formulating polymer, with or without the surfactant TPGS.

The shape and morphology of representative film samples were examined using SEM, as shown in Figure 12. The HPMC-based films (F2 and F4), but not the HPMC-AS-based films (F6 and F8), had submicron silodosin parti-cles and tiny pores (Figure 12). This observation was in agreement with the PXRD results. The submicron particles of silodosin may have formed at the point of supersaturation in HPMC during preparation of the casting gel (148).

43

Figure 12. Scanning electron microscopy (SEM) micrographs of the formulated films and pure silodosin. The bar represents 100 μm for samples F2, F4, F6, and F8 and 20 μm for pure silodosin.

The solid-state characteristics of the films (F1-F8) were investigated using DSC, PXRD and disintegration times. As shown in Figure 13, the melting peak (Tm; 105.97 ˚C) for the crystalline form was absent from all the DSC thermograms of all the prepared films (F1-F8). This could be the result of molecular dispersion of the drug in the polymer.

The fastest disintegration was observed with the HPMC-AS-based film F5 (drug:polymer ratio 1:5) which disintegrated in 15.3±1.2 sec (Table 7). Disintegration was slower for the film based on the HPMC polymer with the same drug:polymer ratio (F1: 35.3±0.6 sec). This effect may have been relat-ed to the properties of the polymer, i.e. wettability and surface tension. Thus, HPMC-AS films disintegrated more quickly than HPMC films (149, 150). Addition of the surfactant (TPGS) to the film formulation had an additive effect on the disintegration time (Table 7). These results were in agreement with those of Vuddanda et al., who found an additive effect on speed of dis-integration on addition of the surfactant TPGS (151).

44

Figure 13. Differential scanning calorimetry of the silodosin film formulations showing (a) HPMC-based films F1-F4 and (b) HPMC-AS-based films F5-F8.

The mechanical and tensile properties of the thin films were measured under ambient conditions. The mechanical properties of HPMC films were better than those of HPMC-AS films with respect to stability during patient handling and packing.

Table 7. Silodosin oral films: Disintegration time

Formulation code Mean disintegration time (seconds)* F1 35.3±0.6 F2 61.0±0.0 F3 65.7±0.6 F4 62.7±1.5 F5 15.3±1.2 F6 33.7±2.5 F7 33.0±1.0 F8 56.7±0.6

*Means ± standard deviations (n ≥ 3)

5.3.1. In vitro dissolution in simulated saliva (Paper IV) The dissolution studies were conducted in simulated saliva (pH 6.8) under non-sink conditions, mimicking the conditions of the oral cavity. The solu-bility of pure silodosin in simulated saliva was 0.46 mg/mL. The solubility of silodosin in simulated saliva with additional dissolved HPMC (with and without TPGS) ranged from 0.48 to 0.51 mg/mL at equilibrium after 48 hours, and with additional dissolved HPMC-AS (with and without TPGS) ranged from 0.94 to 1.1 mg/mL. The dissolution results for the films formulated using HPMC and HPMC-AS are shown in Figure 14a and 14b, respectively. In the HPMC-based films, 80% of the drug was released in 10 minutes from F2 (drug:polymer ratio 1:3), while dissolution was poorer for F1 (drug:polymer ratio 1:5), with 80%

45

release after 25 minutes. This may be because the increase in polymer con-centration led to the formation of a gel-like state which decreased water up-take and retarded drug release, as reported by Singh et al (152). Alhayali et al. have previously reported that, in some cases of solid drug dispersions, the drug concentrations do not change significantly with higher polymer ratios (153).

Addition of TPGS also improved the drug release noticeably. Thus, films containing TPGS dissolved faster than TPGS-free films. Other workers have mentioned this effect of the surfactant TPGS in terms of its ability to im-prove the solubility, dissolution rate and bioavailability of some drugs for-mulated as oral films (154, 155). In HPMC-AS-based films, about 80% of the drug was released in around 10 minutes (F5 and F6); with no significant differences in dissolution rate between the 1:3 and 1:5 ratios. Addition of TPGS to the HPMC-AS-based films negatively affected the dissolution rate for both ratios. Therefore, this study has demonstrated that HPMC works well as a film-forming polymer when combined with TPGS.

Figure 14. Film dissolution of (a) HPMC-based films and (b) HPMC-AS-based films in simulated saliva.

In a case study (156), HPMC was the most suitable film-forming material of the excipients tested, providing faster dissolution and easier-to-handle films. Interestingly, the submicron silodosin particles seen in the HPMC films appear not to have affected the faster drug release from these films. Further, it was observed that the dissolution behavior of HPMC-AS (TPGS-free) films was similar to that of HPMC films containing amorphous solid dispersions (Figure 14). These studies suggested that both polymers are ca-pable of increasing and prolonging the supersaturation of the drug and help-ing to prevent the tendency of the drug to precipitate and recrystallize during dissolution. This could be attributed to the existence of the drug in the amor-phous state and it being molecularly dispersed in the polymer matrix. This was also evident from the thermal, solid-state and morphological results. It has been reported that cellulose-derived polymers such as HPMC, HPMC-AS and HPMCP are superior for preparing solid, amorphous dispersions,

46

particularly with PWSDs, compared to other polymers. The bulky structure of the polymer network and worse solubility properties (compared to highly water-soluble polymers) facilitate the reduction of drug mobility in the pol-ymer matrix and prevent the drug from recrystallizing, as normally induced by supersaturation during dissolution (157). This consequently improves the product’s physical stability and in vitro drug release performance. In general, our dissolution results revealed that either HPMC or HPMC-AS would be useful for preparing films intended for oral cavity absorption, which is more complex than GIT absorption.

Figure 15. Degree of supersaturation as a result of formulating the drug silodosin in different oral film formulations containing HPMC (F1-F4) or HPMC-AS (F5-F8). n= 3; means ± standard deviation.

The DS values for silodosin formulated as an oral film are presented in Figure 15. The highest DS was obtained for HPMC-based formulations at early time points. F4, in particular, had a DS value of > 1 (supersaturated) before 5 minutes, which was earlier than for the other films formulated using the HPMC polymer (F1-F3), as shown in Figure 15. In the dissolution re-sults, 80% of the drug was released from film formulation F4 during the first 10 minutes (Figure 14). This effect could be attributed to a synergistic effect of the surfactant (TPGS) and the polymer (HPMC), resulting in improved solubility and maintained supersaturation (114, 115, 86). Therefore, film formulation F4 appears to have potential for the development of a silodosin film formulation with improved performance and improved oral sublingual absorption as a result of the high degree of supersaturation solubility (10, 158).

47

5.4. Factors affecting supersaturation behavior (Paper I, II, III) 5.4.1. Effects of process, formulation and temperature on supersaturation in FaSSIF (Paper I) The supersaturation and precipitation behavior of solid dispersions of EZ was investigated in a one-compartment model under non-sink conditions. Figure 16 shows the dissolution profiles (concentration versus time) for the prepared solid dispersions in FaSSIF. The solubility of crystalline EZ in FaSSIF (pH 6.5) at 25˚C and 37 ˚C was 5.9 mg/mL and 10.1 mg/mL, respec-tively as shown in Figure 16a and b. The apparent solubility (concentration in solution) of EZ from the amorphous form increased and then rapidly de-creased at 25 ˚C (Figure 16a and b). In contrast, there was no supersaturated concentration at 37 ˚C for amorphous EZ (Figure 16c and d). This phenome-non of supersaturation and rapid precipitation is typical of amorphous solids and is described as the ‘spring’ effect (10).

Figure 16. Dissolution and precipitation (supersaturation) profiles of ternary solid dispersions of ezetimibe (EZ) in fasted-state simulated intestinal fluid. (a) Melt-quenched (MQ1, 2, 3) and (b) spray-dried (SD1, 2, 3) solid dispersions at 25˚C; (c) MQ and (d) SD solid dispersions at 37˚C. AMP: amorphous EZ; CRY: crystalline EZ.

48

For orally administered formulations, the dose-to-solubility ratio has been suggested as an indicator of the precipitation propensity of the drug in the intestinal lumen (6). Drugs with dose-to-solubility (non-ionized) ratios of more than 50:1 have been reported to show a tendency to precipitate in the intestinal environment (159). The dose-to-solubility ratio for EZ is about 1000 at a dose of 10 mg, suggesting that the observed precipitation of stable phases is thermodynamically favored in FaSSIF. Interestingly, the ‘spring’ effect disappeared (i.e. there was no increase in solubility) as the temperature of the dissolution experiment was increased to 37 ˚C. This may have been because of differences in the relationship between dissolution and the precip-itation/crystallization kinetics of crystalline EZ. It is possible that the crystal-lization of amorphous EZ was faster at 37 ˚C. Similar behavior has been observed with amorphous felodipine and indomethacin in another study (160).

Solid dispersions where the drug and polymers are intimately mixed are widely used formulations that can prevent the crystallization of the stable phase and maintain the supersaturation of drugs like EZ. In this study, PVP-K30 and poloxomer 188 were selected from the preliminary studies as the precipitation inhibitor and solubilizer, respectively. PVP has been well estab-lished as a crystal growth inhibitor for a range of amorphous drugs (126).

In the MQ solid dispersions, concentrations of EZ in solution were main-tained at about 50-55 mg/mL and no precipitation was observed for at least 240 min at 25 ˚C (Figure 16a). Apparently, the presence of the polymer sus-tained supersaturation, with solution concentrations about 50 times higher than the equilibrium solubility of EZ, possibly by preventing precipitation of the stable phase. However, no differences in the solution concentrations were observed for different drug:polymer ratios in these solid dispersions.

Interestingly, in the SD solid dispersions, supersaturation (with solution concentrations about 70-80 times higher) was followed by desupersaturation at different rates (Figure 16b). SD solid dispersions generated higher initial solution concentrations than those generated by MQ solid dispersions, which could be explained by differences in the particle size/morphology and mo-lecular miscibility. This could also mean that solid dispersions result in a greater driving force for crystallization. The desupersaturation curves also showed that the lower the proportion of polymer in the solid dispersions, the faster the precipitation kinetics will be (Figure 16b).

The presence of 10% poloxomer resulted in even faster precipitation of the crystalline phase. The rich drug–polymer interactions and changes in viscosity can alter the crystal growth and surface integration kinetics, leading to differences in the crystallization kinetics. Previously published articles have also proposed similar mechanisms to explain the differences in the precipitation kinetics of many amorphous drugs (103).

49

In general, solid dispersions prepared by the melting method showed slightly higher initial concentrations and faster precipitation kinetics at 37 ˚C than at 25 ˚C (Figure 16a and c), while solid dispersions prepared by the spray-drying method showed direct precipitation in FaSSIF at 37 ˚C but not at 25 ˚C (Figure 16b and d). The presence of precipitation inhibitors had only a minor effect in maintaining supersaturation at 37 ˚C. These results are per-fectly in line with previous observations on the supersaturation and precipi-tation behavior of other amorphous drugs in the presence of different poly-mers (161). The results suggest that dissolution experiments performed at 25 ˚C may not be biorelevant and may fail to predict in vivo behavior (162).

In the MQ solid dispersions, desupersaturation was dependent on the for-mulation: a higher polymer/poloxomer content resulted in a slower precipita-tion rate (Figure 16c). However, in the SD solid dispersions, precipitation was rapid irrespective of the formulation or polymer content (Figure 16d). This suggests that the supersaturation and precipitation behavior of amor-phous solids is sensitive to the processing method, possibly because of the history of the material and the presence of nuclei (163).

5.4.2. Effects of process, formulation and temperature on supersaturation in FaSSGF (Paper I) Figure 17 shows the dissolution (concentration versus time) profiles for EZ samples in FaSSGF (pH 1.6). The solubility of crystalline EZ in FaSSGF (pH 1.2) at 25 ˚C and 37 ˚C was extremely low: 0.2 mg/mL and 0.1 mg/mL, respectively (Figure 17a and b). This extremely low solubility is thought to be due to the un-ionized state of the drug in the dissolution medium. In con-trast to its behavior in FaSSIF, there was no increase in initial solution con-centrations or supersaturation for amorphous EZ in FaSSGF. The method of preparation and the temperature had no effect on this behavior. This suggests rapid transformation of the amorphous phase into the crystalline phase, pos-sibly as the result of an extremely high driving force for crystallization.

Although some increase in the initial EZ concentrations and supersatura-tion were observed with the MQ and SD solid dispersions in general, the level and extent of supersaturation were much lower in FaSSGF than in FaSSIF at 25 ˚C and 37 ˚C (Figure 17). Higher polymer/poloxomer content delayed the precipitation kinetics, leading to sustained supersaturation for longer periods as observed in FaSSIF.

Although PVP and poloxamer 188 were effective in inhibiting/delaying precipitation from the solid dispersions at 25 ˚C, they were ineffective at 37 ˚C, which is the physiologically relevant temperature. The method of prepa-ration also affected the supersaturation and precipitation kinetics.

In general, supersaturation was maintained for relatively longer in bio-relevant media at both temperatures when MQ solid dispersions were used

50

(Figure 17). It is important to consider these aspects in the design and devel-opment of amorphous solid dispersions of poorly soluble drugs.

Figure 17. Dissolution and precipitation (supersaturation) profiles of ternary solid dispersions of ezetimibe (EZ) in fasted-state simulated gastric fluid. (a) Melt-quenched (MQ1, 2, 3) and (b) spray-dried (SD1, 2, 3) solid dispersions at 25 ºC; (c) MQ and (d) SD solid dispersions at 37 ºC. AMP: amorphous EZ, CRY: crystalline EZ.

5.4.3. Effect of implementation of an absorptive compartment on supersaturation behavior (Paper II) The absorption process across the intestinal epithelial barrier could substan-tially influence the supersaturation and precipitation behavior of a drug. In this study, the effect of absorption across Caco-2 cell monolayers on the supersaturation and precipitation of EZ was investigated in a cell-free, one-compartment setup and compared with results in a two-compartment model.

Supersaturation experiments were initially conducted in wells of cell cul-ture plates (one-compartment setup) to gain insight into the supersatura-tion/precipitation behavior of EZ in the absence of Caco-2 cells (i.e. in a cell-free environment).

As illustrated in Figure 18a, the EZ concentration was higher with the DS10 sample than with crystalline EZ, and supersaturation was maintained until the end of the experiment. EZ precipitated with the DS20 and DS30 samples after 15 and 5 min, respectively. Instantaneous precipitation oc-curred with the DS40 sample in FaSSIF (Figure 18a). In this study, sponta-

51

neous nucleation and precipitation of EZ were observed in FaSSIF for all the supersaturated solutions, suggesting a low kinetic barrier to crystallization of the drug. This is typical behavior, because the supersaturated state is ther-modynamically unstable and faster precipitation occurs as supersaturation increases, according to the theory of crystal growth (126). In contrast, Bevernage et al. observed no spontaneous precipitation, and nuclei were required to promote precipitation (164).

Figure 18. (a) Amount of ezetimibe (EZ) in fasted-state simulated intestinal fluid (FaSSIF) in the one-compartment setup. (b) Amount of EZ in FaSSIF on the apical side of Caco-2 cell monolayers as a function of time, after induction of supersatura-tion. (c) Cumulative amount of EZ recovered from the basolateral acceptor medium (i.e. HBSS) after vectorial transport across Caco-2 cell monolayers upon induction of supersaturation in the apical compartment. Data represent means ± standard deviations (n=3). Error bars in (c) were too small to show. DS = degree of super-saturation.

Figure 18b shows the total amounts of EZ in FaSSIF on the AP side (two-compartment model) without addition of polymer. As depicted, the amount of drug dissolved increased as the DS increased. All DS levels were associ-ated with more stable solutions with lower precipitation tendencies. The reason for this could be that permeation of the drug was an alternative to precipitation for relieving the thermodynamically unstable state that is the predominant feature during supersaturation, and as the concentration (or the

52

DS) increased, the possibility of permeation reducing precipitation also in-creased, as reported by Bevernage et al. .

The effect of inducing different degrees of supersaturation on the permea-tion of EZ across the Caco-2 cell monolayer is shown in Figure 18c. The cumulative amounts of EZ in the transport medium (HBSS) in the BL com-partment is plotted as a function of time. As delineated in Figure 18c, the transport of drug to the BL side was delayed for about 45 min with DS10 and 30 min with DS20, DS30 and DS40. This lag-time could be due to the time needed to saturate the monolayer barrier before reaching steady-state flux, as reported by Kanzer et al. (165).

The highest amount of EZ transported to the BL compartment after 2 h was 1.28 ± 1.1 ×10-4 μg from the DS40 sample; 40.83 ± 2.27 μg was in solu-tion in the AP compartment after 2 h, as shown in Figure 18b. The resulting flux was 0.012 ± 9.9 ×10-4 μg/min×cm2 (Figure 19). Although the DS10 sample resulted in the highest amount of EZ dissolved after 2 h in the one-compartment setup (Figure 18a), the highest amount transported after 2 h in the two-compartment setup was from the DS40 sample. The DS40 sample also maintained the highest amount of dissolved drug in the supersaturated state on the AP side of the two-compartment setup (Figure 18a and b).

Figure 19. Calculated flux of various ezetimibe (EZ) formulations across Transwell-grown Caco-2 cell monolayers. * Indicates a significant statis-tical difference in the flux between the different polymer concentrations (p < 0.05), and ns indicates that there were no statistically significant differences in flux beween the indicated samples (with varying degrees of supersaturation; DS) (p > 0.05).

53

Differences between the DS samples in the amounts of EZ transported to the BL side were less pronounced than differences in the precipitation behavior in the one-compartment setup. However, precipitation behavior in the one-compartment setup was not in accordance with that on the AP side in the two-compartment setup. The different outcomes for the one-compartment and two-compartment setups demonstrate that reliable prediction of super-saturation/precipitation behavior cannot be achieved without including tools to predict absorption in the in vitro setup, as reported by Bevernage et al. (164). As shown in Figure 19, there were no significant differences in flux values between samples with DS10, DS20, DS30 or DS40 for drug transport to the BL side (p > 0.05). However, the EZ fluxes with DS10, DS20, DS30 and DS40 samples differed significantly from the flux with the crystalline EZ sample (p < 0.05).

As reported in a previous study (166), the amount of drug transported to the BL side depends on the amount of drug in the supersaturated solution on the AP side. The specific behavior of EZ as it was taken up and metabolized by enterocytes could explain the lack of effect of the various DSs on EZ flux on the apical side of the two-compartment setup, as reported by Oswald et al. (167). Generally, PWSDs that permeate the intestinal wall well should show increased permeation when the concentration of the dissolved drug is in-creased on the AP side, but some recent reports have shown that solubilized colloidal drugs may permeate poorly (166, 168).

5.4.4. Effect of polymer on supersaturation-precipitation-permeation interplay (Paper II) PVP K-30 was selected to assess the effects of polymers on the inhibition of crystallization (i.e. on sustaining supersaturation) and transport (169). In the one-compartment setup with the DS40 sample (Figure 20a), a PVP K-30 concentration of 0.05% w/v resulted in a higher amount of drug in solution but precipitation occurred gradually over 90 min, at which point all the drug had been precipitated. A PVP K-30 concentration of 0.1% w/v resulted in rapid precipitation over the first 30 min; i.e. it did not perform better than 0.05% w/v in keeping the supersaturation state. Precipitation occurred later with both polymer concentrations (0.05% and 0.1% w/v) than in the absence of polymer.

In the two-compartment setup, a concentration of 0.05% w/v maintained the supersaturation for 2 h (Figure 20b), but it was difficult to conclude the effect of 0.1% w/v because of system variability. In general, less extensive precipitation was observed in the two-compartment setup, and the amount of EZ dissolved on the AP side was greater in the presence of PVP K-30 (Fig-ure 20b). The polymer PVP K-30 (0.05% w/v) kept the drug in the supersat-urated state for 2 h. In the absence of polymer, the drug started to precipitate

54

after 15 min. On the other hand, PVP K-30 (0.1% w/v) in the medium on the AP side increased the amount of dissolved drug up to 20 μg after 5 min, which was more than with a concentration of 0.05% w/v and in the absence of polymer (Figure 20b).

The effect of PVP K-30 on the supersaturation of EZ in FaSSIF has been demonstrated previously (153). The reason for less extensive precipitation of EZ in the presence of Caco-2 cells may be that permeation offered an alternative to precipitation, as discussed earlier. The same conclusion applies in the absence of polymer. Thus, the in vitro prediction of supersaturation and precipitation is more accurate if it is combined with absorption tools because, in cases like this, the precipitation can be overestimated by the one-compartment experiments compared to the two-compartment experiments.

Figure 20. (a) Amount of ezetimibe (EZ) with PVP-K30 in fasted-state simulated intestinal fluid (FaSSIF) in the one-compartment setup. (b) Amount of EZ with PVP-K30 in FaSSIF on the apical side of Caco-2 cell monolayers as a function of time, after nduction of supersaturation to DS40. (c) Cumulative amount of EZ recovered from the basolateral acceptor medum (i.e. HBSS) after vectorial transport across Caco-2 cell monolayers upon induction of supersaturation in the presence of poly-mer in the apical compartment. Data represent means ± standard deviation (n=3). Error bars in (c) were too small to show. DS=degree of supersaturation.

In this study, we also investigated how supersaturation affects the transport of EZ in vitro in a two-compartment setup using Caco2 cells in the presence of PVP K-30. Figure 20c shows the cumulative amounts of EZ on the BL

55

side that were transported over 2 h. Addition of 0.05% w/v PVP K-30 to the AP side resulted in the highest amount of EZ transported to the BL side. Furthermore, the flux of EZ was significantly improved (p < 0.05) with this polymer concentration (0.064 ± 1.5 x10-2 μg/min×cm compared to 0.1% w/v polymer and no polymer; Figure 19). PVP K-30 (0.1% w/v) improved the supersaturation solubility of EZ on the apical side in FaSSIF without im-proving the flux; the total transported amount was 2.97 ± 7.5 x 10-4 μg. In summary, PVP K-30 (0.05% w/v) on the AP side in FaSSIF kept the drug in the supersaturated state for 2 h, improved the flux, and achieved the highest cumulative EZ concentration on the BL side after 2 h (5.54 μg ± 2.6 x10-3), as shown in Figure 20b and c. The increase in drug concentration on the donor AP side as a result of solubilizing the excipients may not cause a pro-portional improvement in the flux in all cases (170). The intrinsic physico-chemical properties of the drug and drug-excipient interaction can be ex-pected to affect the outcome. In this study, PVP K-30 improved the flux across the Caco-2 cell barrier (Figure 19). However, some recent studies have shown the opposite effect, reporting that addition of some excipients to the AP side has a negative effect on transport. This was attributed to the fact that increased apparent solubility is associated with a micellar solubilization effect on the excipients which can result in overestimation of the true super-saturation, thus not causing improvement in the permeation rate (166).

5.4.5. Effect of polymer on the supersaturation behavior of cocrystals (Paper III) The polymer HPMC was used as a model in this study. In the presence of 0.1% w/v HPMC, cocrystal and amorphous forms achieved much higher concentrations than the crystalline drug form, and the supersaturation was sustained for longer time compared to dissolution in the absence of HPMC (Figure 11). The dissolution of TDF-MOA cocrystals in the presence of 0.1% w/v HPMC resulted in a Cmax of ∼0.9× 10-4 M, which corresponds to a supersaturation (Cmax/Sdrug) of about 120. The amorphous form was associat-ed with similar concentrations. For both cocrystals and the amorphous phase, supersaturation was maintained much longer (3 hours) in the presence of HPMC; i.e., this suggests slower conversion of the amorphous or cocrystal forms to the drug. HPMC at a concentration of 0.1% w/v provided maximum precipitation inhibition from a solution supersaturated at 29 (171). These findings suggest that cocrystals may increase drug bioavailability under similar conditions.

56

57

6. ConclusionsThis thesis has provided a thorough understanding of the in vitro supersatu-ration solubility behavior of some newly formulated supersaturating SDDSs such as amorphous solid dispersions, cocrystals and oral films. During the course of this thesis:

• Ternary solid dispersions of EZ containing PVP and poloxamer 188were successfully prepared using spray-drying and melt-quenchingmethods. The powders were thoroughly characterized using solid-statetools and were found to be predominantly amorphous. Cocrystals ofTDF with MOA were identified and thoroughly characterized. Sublin-gual oral films of silodosin were prepared successfully for the first time.HPMC-AS polymer-based films showed fully amorphous dispersion ascompared to HPMC-based films.

• It was found that MQ solid dispersions had better supersaturation prop-erties than SD dispersions. The melting method would thus seem to bemore promising for enhancing the dissolution of EZ from a ternary soliddispersion. The apparent solubility of EZ solid dispersions was higher inFaSSIF than in FaSSGF and supersaturation was maintained for an ade-quate time, i.e. for the time required for the absorption process to takeplace. The precipitation kinetics of EZ in solid dispersions were faster at37 ˚C than at 25 ˚C in biorelevant media, indicating that dissolutionstudies currently routinely performed at 25 ˚C may be less relevant.TDF-MOA cocrystal solubility was higher than that of TDF over a widepH range (1-7) in aqueous media. The dissolution of cocrystals created asupersaturation level that was significantly higher than the saturationpoint of the crystalline form. The SA of TDF-MOA cocrystals increasedby several orders of magnitude with increases in pH.

• The supersaturated solutions of EZ showed less precipitation tendency inthe presence of Caco-2 cells in a two-compartment system and higherprecipitation at early time points in a one-compartment chamber. There-fore, the supersaturation studies that are usually conducted in one-compartment chambers or in a test tube could underestimate supersatura-tion of the drug. Implementation of an absorption tool (such as Caco-2cell monolayers) in supersaturation evaluation tests in vitro could im-prove supersaturation and precipitation predictions.

58

• The effects of polymer and surfactant excipients on the supersaturation and precipitation behavior of SDDs differed. Low concentrations of poloxamer 188 were adequate for obtaining the desired improvements in EZ solubility in ternary solid dispersions. The polymer PVP K-30 im-proved the supersaturation of EZ at concentrations of 0.05% w/v and re-sulted in significant improvements in the transport/flux (J). The addition of 0.1% w/v HPMC to the dissolution medium increased supersaturation and slowed down nucleation and/or precipitation kinetics of the drug from formulations of TDF-MOA cocrystals and the amorphous form. The dissolution behavior of films containing HPMC is similar to that of those containing HPMC-AS if the surfactant TPGS 0.5 % w/w is included in the film formulation.

59

7. Future workFuture work could include:

˗ Investigation of other factors possibly affecting supersaturation be-havior (such as hydrodynamics and modified biorelevant media) and carrying out online investigation of the precipitation kinetics.

˗ Investigation of the supersaturation-precipitation-permeation inter-play of EZ in a Caco-2 cell environment for a solid dispersion for-mulation of the drug.

˗ Investigation of the supersaturation behavior of TDF cocrystals in various biorelevant media and aspirated human intestinal fluids.

˗ Investigation of the in vivo behavior of the prepared formulations, since only in vitro dissolution tests and cell culture studies were per-formed in this thesis.

˗ Investigation of improvements to the formulation of the silodosin films, since they were prepared for the first time in this thesis. For example, studies of the dissolution of the films in human saliva could be necessary to evaluate their fate in realistic conditions. Per-meability investigations are important for evaluating film absorption behavior, and film palatability and crystallization of the drug during storage (stability) also require investigation.

60

61

8. AcknowledgmentsForemost I would like to thank Allah for giving me the power and energy to be here defending my thesis, and for helping me to overcome my weakness-es during the hardest times.

The studies in this thesis were carried out at the Department of Health Sci-ences, Division of Medical Sciences, at Luleå University of Technology. Work was done within the Pharmaceutical and Biomaterial research group. The financial support I received was from the Ministry of Higher Education and Scientific research in Iraq for PhD study.

I wish to express my sincere gratitude to: My main supervisor: Professor Sitaram Velaga, for his guidance and support throughout the course of my thesis, and also for his valuable input during meetings and discussions throughout my journey through the research work.

My former co-supervisor Staffan Tavelin for his guidance during research work for Papers I and II and for hosting me in his laboratory at Umeå Uni-versity, Sweden.

My new co-supervisor Parameswara Vuddanda for his help and support per-forming the research work in Paper IV.

Prof. Carsten Ehrhardt and his group members for hosting me at their labora-tory in the School of Pharmacy and Pharmaceutical Sciences and Trinity Biomedical Sciences Institute, Trinity College Dublin, Dublin, Ireland.

Prof. Sung-Joo Hwang and his group members for their support during my visit to their laboratory in Yonsei University, South Korea.

I would also like to sincerely and gratefully thank my colleagues and friends in the Department of Health Sciences and Division of Medical Sciences for their patience and for being my second work family here in Luleå. Special thanks go to the members of the administration department for helping me to improve my Swedish language mainly; Katarina Virtoft. Thank you Niklas Lehto for your help with the thesis. Many thanks to Katarina Leijon-Sundqvist for encouragement. Many thanks to my former colleague Hassan Ali for his advice and my new colleague Mohammed Elbadawi for support and encouragement. I won’t forget that you told me “Amani, we are not only

62

pharmacists but also engineers” when we managed to solve technical prob-lems in the laboratory.

Great, heartfelt thanks go to my friend Manar for believing in me: you are really a true friend who had my back at all times. Thanks to all the friends I met in Luleå: Eynas, Asma, Hana, Hiba, Sara, Zahraa – we had a really fun time and spent some unforgettable moments laughing and enjoying the win-ter in the snow here in the north of Sweden. Special thanks to my friends from Egypt: mainly Taisir, Heba, Noha, Basma. You made my time here in Luleå different and gave me constant support to enable me to reach this day. Thanks for your help as my second family during my travel to Dublin, leg surgery and during the days of writing my thesis.

I wish to express my deep gratitude to my dearest parents for their encour-agement, love and prayers for me throughout my time here. Thank you, you taught me how to be who I want to be, how to forgive, and how to be inde-pendent. Gratitude is also extended to my brothers and sisters for all the support they have given to me.

My deepest gratitude goes to my husband Hussan for his support and for being so patient during my research, to my beloved son Toto for taking care of me during my illness and making me feel that I am a very important per-son in his life, and to my son Yaman (the chef) who cooked our food when I was busy with the thesis. You have been the driving force for my carrying out this work.

Amani Alhayali June, 2018 Luleå, Sweden

63

9. References1. Lawrence XY, Lipka E, Crison JR, Amidon GL. Transport approaches to the biopharma-

ceutical design of oral drug delivery systems: prediction of intestinal absorption. AdvDrug Deliv Rev. 1996;19(3):359-76.

2. Löbenberg R, Amidon GL. Modern bioavailability, bioequivalence and biopharmaceuticsclassification system. New scientific approaches to international regulatory standards.Eur J Pharm Biopharm. 2000;50(1):3-12.

3. Zhang GG, Law D, Schmitt EA, Qiu Y. Phase transformation considerations during processdevelopment and manufacture of solid oral dosage forms. Adv Drug Deliv Rev.2004;56(3):371-90.

4. Cui Y. A material science perspective of pharmaceutical solids. Int J Pharm. 2007;339(1-2):3-18.

5. Chow S. Bioavailability and bioequivalence in drug development. Wiley InterdisciplinaryReviews: Comput Stat. 2014;6(4):304-12.

6. Amidon GL, Lennernäs H, Shah VP, Crison JR. A theoretical basis for a biopharmaceuticdrug classification: the correlation of in vitro drug product dissolution and in vivo bioa-vailability. Pharm Res. 1995;12(3):413-20.

7. Bevernage J, Brouwers J, Annaert P, Augustijns P. Drug precipitation–permeation inter-play: supersaturation in an absorptive environment. Eur J Pharm Biopharm.2012;82(2):424-8.

8. Bikiaris DN. Solid dispersions, part I: recent evolutions and future opportunities in manu-facturing methods for dissolution rate enhancement of poorly water-soluble drugs. ExpertOpin Drug Deliv. 2011;8(11):1501-19.

9. Singh A, Worku ZA, Van den Mooter G. Oral formulation strategies to improve solubilityof poorly water-soluble drugs. Expert Opin Drug Deliv. 2011;8(10):1361-78.

10. Brouwers J, Brewster ME, Augustijns P. Supersaturating drug delivery systems: Theanswer to solubility-limited oral bioavailability? J Pharm Sci. 2009; 98(8):2549-72.

11. Gao P, Shi Y. Characterization of supersaturatable formulations for improved absorptionof poorly soluble drugs. The AAPS journal. 2012;14(4):703-13.

12. Martinez MN, Amidon GL. A mechanistic approach to understanding the factors affectingdrug absorption: a review of fundamentals. J Clin Pharmacol. 2002;42(6):620-43.

13. Benet LZ, Wu C, Hebert MF, Wacher VJ. Intestinal drug metabolism and antitransportprocesses: A potential paradigm shift in oral drug delivery. J Control Release. 1996 May1996;39(2):139-43.

14. Macheras P, Karalis V. A non-binary biopharmaceutical classification of drugs: The ABΓsystem. Int J Pharm. 2014;464(1-2):85-90.

15. Lipinski CA. Drug-like properties and the causes of poor solubility and poor permeability.J Pharmacol Toxicol Methods. 2000;44(1):235-49.

16. McNaught AD, Wilkinson A. International Union of Pure and Applied Chemistry. Com-pendium of chemical terminology: IUPAC recommendations. 1997; 2.

17. Hoshino T, Uekama K, Pitha J. Increase in temperature enhances solubility of drugs inaqueous solutions of hydroxypropylcyclodextrins. Int J Pharm. 1993; 98(1-3):239-42.

18. Garzón LC, Martínez F. Temperature dependence of solubility for ibuprofen in someorganic and aqueous solvents. J Solution Chem. 2004;33(11):1379-95.

19. Stegemann S, Leveiller F, Franchi D, De Jong H, Lindén H. When poor solubility be-comes an issue: from early stage to proof of concept. Eur J Pharm Sci. 2007; 31(5):249-61.

64

20. Vasconcelos T, Sarmento B, Costa P. Solid dispersions as strategy to improve oral bioa-vailability of poor water soluble drugs. Drug Discov Today. 2007; 12(23-24):1068-75.

21. Lipinski CA, Lombardo F, Dominy BW, Feeney PJ. Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings1. Adv Drug Deliv Rev. 2001; 46(1):3-26.

22. Frank KJ, Rosenblatt KM, Westedt U, Hölig P, Rosenberg J, Mägerlein M, et al. Amor-phous solid dispersion enhances permeation of poorly soluble ABT-102: true supersatu-ration vs. apparent solubility enhancement. Int J Pharm. 2012;437(1-2):288-93.

23. Jornada DH, dos Santos Fernandes, Guilherme Felipe, Chiba DE, de Melo, Thais Regina Ferreira, dos Santos JL, Chung MC. The prodrug approach: a successful tool for improv-ing drug solubility. Molecules. 2015;21(1):42.

24. Nelson E. Solution rate of theophylline salts and effects from oral administration. J Pharm Sci. 1957;46(10):607-14.

25. Nelson E. Comparative dissolution rates of weak acids and their sodium salts. J Am Pharm Assoc. (Scientific ed.). 1958;47(4):297-9.

26. Noyes AA, Whitney WR. The rate of solution of solid substances in their own solutions. J Am Chem Soc. 1897;19(12):930-4.

27. Leuner C, Dressman J. Improving drug solubility for oral delivery using solid dispersions. Eur J Pharm Biopharm. 2000;50(1):47-60.

28. Williams HD, Trevaskis NL, Charman SA, Shanker RM, Charman WN, Pouton CW, et al. Strategies to address low drug solubility in discovery and development. Pharmacol Rev. 2013 Jan;65(1):315-499.

29. Sun J, Wang F, Sui Y, She Z, Zhai W, Wang C, et al. Effect of particle size on solubility, dissolution rate, and oral bioavailability: evaluation using coenzyme Q(1)(0) as naked nanocrystals. Int J Nanomedicine. 2012;7:5733-44.

30. Blagden N, de Matas M, Gavan PT, York P. Crystal engineering of active pharmaceutical ingredients to improve solubility and dissolution rates. Adv Drug Deliv Rev. 2007;59(7):617-30.

31. Vogt M, Kunath K, Dressman JB. Dissolution enhancement of fenofibrate by microniza-tion, cogrinding and spray-drying: comparison with commercial preparations. Eur J Pharm Biopharm. 2008;68(2):283-8.

32. Kawakami K, Pikal MJ. Calorimetric investigation of the structural relaxation of amor-phous materials: Evaluating validity of the methodologies. J Pharm Sci. 2005;94(5):948-65.

33. Stillinger FH. Supercooled liquids, glass transitions, and the Kauzmann paradox. J Chem Phys. 1988;88(12):7818-25.

34. Alhayali A, Tavellin S, Velaga S. Dissolution and precipitation behavior of ternary solid dispersions of ezetimibe in biorelevant media. Drug Dev Ind Pharm. 2017;43(1):79-88.

35. Ambike AA, Mahadik K, Paradkar A. Spray-dried amorphous solid dispersions of simvas-tatin, a low T g drug: in vitro and in vivo evaluations. Pharm Res. 2005; 22(6):990-8.

36. Pokharkar VB, Mandpe LP, Padamwar MN, Ambike AA, Mahadik KR, Paradkar A. Development, characterization and stabilization of amorphous form of a low Tg drug. Powder Technol. 2006;167(1):20-5.

37. Sekiguchi K, Obi N. Studies on Absorption of Eutectic Mixture. I. A Comparison of the Behavior of Eutectic Mixture of Sulfathiazole and that of Ordinary Sulfathiazole in Man. Chem Pharm Bull. 1961;9(11):866-72.

38. Kanig JL. Properties of fused mannitol in compressed tablets. J Pharm Sci. 1964;53(2):188-92.

39. Simonelli A, Mehta S, Higuchi W. Dissolution rates of high energy polyvinylpyrrolidone (PVP)‐sulfathiazole coprecipitates. J Pharm Sci. 1969;58(5):538-49.

65

40. Ohara T, Kitamura S, Kitagawa T, Terada K. Dissolution mechanism of poorly water-soluble drug from extended release solid dispersion system with ethylcellulose and hy-droxypropylmethylcellulose. Int J Pharm. 2005;302(1-2):95-102.

41. Van Drooge D, Hinrichs W, Visser M, Frijlink H. Characterization of the molecular dis-tribution of drugs in glassy solid dispersions at the nano-meter scale, using differential scanning calorimetry and gravimetric water vapour sorption techniques. Int J Pharm. 2006;310(1-2):220-9.

42. Majerik V, Charbit G, Badens E, Horváth G, Szokonya L, Bosc N, et al. Bioavailability enhancement of an active substance by supercritical antisolvent precipitation. J Supercrit Fluids. 2007;40(1):101-10.

43. Karataş A, Yüksel N, Baykara T. Improved solubility and dissolution rate of piroxicam using gelucire 44/14 and labrasol. II Farmaco. 2005;60(9):777-82.

44. Chauhan B, Shimpi S, Paradkar A. Preparation and evaluation of glibenclamide-polyglycolized glycerides solid dispersions with silicon dioxide by spray drying tech-nique. Eur J Pharm Sci. 2005;26(2):219-30.

45. Li F, Hu J, Deng J, Su H, Xu S, Liu J. In vitro controlled release of sodium ferulate from Compritol 888 ATO-based matrix tablets. Int J Pharm. 2006;324(2):152-7.

46. Won D, Kim M, Lee S, Park J, Hwang S. Improved physicochemical characteristics of felodipine solid dispersion particles by supercritical anti-solvent precipitation process. Int J Pharm. 2005;301(1-2):199-208.

47. Dannenfelser R, He H, Joshi Y, Bateman S, Serajuddin AT. Development of clinical dosage forms for a poorly water soluble drug I: Application of polyethylene glycol–polysorbate 80 solid dispersion carrier system. J Pharm Sci. 2004;93(5):1165-75.

48. Craig DQ. The mechanisms of drug release from solid dispersions in water-soluble poly-mers. Int J Pharm. 2002;231(2):131-44.

49. Sugimoto M, Okagaki T, Narisawa S, Koida Y, Nakajima K. Improvement of dissolution characteristics and bioavailability of poorly water-soluble drugs by novel cogrinding method using water-soluble polymer. Int J Pharm. 1998;160(1):11-9.

50. Sarode AL, Wang P, Obara S, Worthen DR. Supersaturation, nucleation, and crystal growth during single-and biphasic dissolution of amorphous solid dispersions: Polymer effects and implications for oral bioavailability enhancement of poorly water soluble drugs. Eur J Pharm Biopharm. 2014;86(3):351-60.

51. Karavas E, Ktistis G, Xenakis A, Georgarakis E. Effect of hydrogen bonding interactions on the release mechanism of felodipine from nanodispersions with polyvinylpyrrolidone. Eur J Pharm Biopharm. 2006;63(2):103-14.

52. Desai J, Alexander K, Riga A. Characterization of polymeric dispersions of dimenhydri-nate in ethyl cellulose for controlled release. Int J Pharm. 2006;308(1-2):115-23.

53. Konno H, Taylor LS. Influence of different polymers on the crystallization tendency of molecularly dispersed amorphous felodipine. J Pharm Sci. 2006;95(12):2692-705.

54. Verreck G, Decorte A, Heymans K, Adriaensen J, Liu D, Tomasko D, et al. Hot stage extrusion of p-amino salicylic acid with EC using CO2 as a temporary plasticizer. Int J Pharm. 2006;327(1-2):45-50.

55. Tanaka N, Imai K, Okimoto K, Ueda S, Tokunaga Y, Ohike A, et al. Development of novel sustained-release system, disintegration-controlled matrix tablet (DCMT) with sol-id dispersion granules of nilvadipine. J Controlled Release. 2005;108(2-3):386-95.

56. García-Zubiri ÍX, González-Gaitano G, Isasi JR. Thermal stability of solid dispersions of naphthalene derivatives with β-cyclodextrin and β-cyclodextrin polymers. Thermo-chimica acta. 2006;444(1):57-64.

57. Karavas E, Georgarakis E, Bikiaris D. Application of PVP/HPMC miscible blends with enhanced mucoadhesive properties for adjusting drug release in predictable pulsatile chronotherapeutics. Eur J Pharm Biopharm. 2006;64(1):115-26.

66

58. Guyot M, Fawaz F, Bildet J, Bonini F, Lagueny A. Physicochemical characterization and dissolution of norfloxacin/cyclodextrin inclusion compounds and PEG solid dispersions. Int J Pharm. 1995;123(1):53-63.

59. Ceballos A, Cirri M, Maestrelli F, Corti G, Mura P. Influence of formulation and process variables on in vitro release of theophylline from directly-compressed Eudragit matrix tablets. II Farmaco. 2005;60(11-12):913-8.

60. Lim W, Lawrence M. Influence of surfactant and lipid chain length on the solubilisation of phosphatidylcholine vesicles by micelles comprised of polyoxyethylene sorbitan mo-noesters. Colloids Surf Physicochem Eng Aspects. 2004;250(1-3):449-57.

61. Pouton CW. Formulation of poorly water-soluble drugs for oral administration: physico-chemical and physiological issues and the lipid formulation classification system. Eur J Pharm Sci. 2006;29(3-4):278-87.

62. Mashru R, Sutariya V, Sankalia M, Parikh P. Development and evaluation of fast-dissolving film of salbutamol sulphate. Drug Dev Ind Pharm. 2005;31(1):25-34.

63. Slavkova M, Breitkreutz J. Orodispersible drug formulations for children and elderly. Eur J Pharm Sci. 2015; 75:2-9.

64. Preis M, Woertz C, Kleinebudde P, Breitkreutz J. Oromucosal film preparations: classifi-cation and characterization methods. Expert Opin Drug Deliv. 2013;10(9):1303-17.

65. Joshua JM, Hari R, Jyothish FK, Surendran SA. Fast Dissolving Oral Thin Films: An Effective Dosage Form for Quick Releases. Drugs. 2016;11:12.

66. Borges AF, Silva C, Coelho JF, Simões S. Oral films: current status and future perspec-tives: I—galenical development and quality attributes. J Controlled Release. 2015;206:1-19.

67. Borges JG, Carvalho RA. Orally disintegrating films containing propolis: properties and release profile. J Pharm Sci. 2015;104(4):1431-9.

68. Bhyan B, Jangra S, Kaur M, Singh H. Orally fast dissolving films: innovations in formula-tion and technology. Int J Pharm Sci Rev Res. 2011;9(2):9-15.

69. Bala R, Pawar P, Khanna S, Arora S. Orally dissolving strips: A new approach to oral drug delivery system. Int J Pharm Investig. 2013 Apr;3(2):67-76.

70. Irfan M, Rabel S, Bukhtar Q, Qadir MI, Jabeen F, Khan A. Orally disintegrating films: a modern expansion in drug delivery system. Saudi Pharm J. 2016;24(5):537-46.

71. Siddiqui MN, Garg G, Sharma PK. A short review on-a novel approach in oral fast dis-solving drug delivery system and their patents. Adv Biol Res. 2011;5(6):291-303.

72. Lee Y, Kim K, Kim M, Choi DH, Jeong SH. Orally disintegrating films focusing on formulation, manufacturing process, and characterization. J Pharm Investig. 2017;47(3):183-201.

73. Zhang H, Zhang J, Streisand JB. Oral mucosal drug delivery. Clin Pharmacokinet. 2002;41(9):661-80.

74. Aitipamula S, Banerjee R, Bansal AK, Biradha K, Cheney ML, Choudhury AR, et al. Polymorphs, salts, and cocrystals: what’s in a name? Cryst Growth Des. 2012;12(5):2147-52.

75. Good DJ, Rodríguez-Hornedo N. Solubility advantage of pharmaceutical cocrystals. Cryst Growth Des. 2009;9(5):2252-64.

76. Friščić T, Jones W. Benefits of cocrystallisation in pharmaceutical materials science: an update. J Pharm Pharmacol. 2010;62(11):1547-59.

77. Gao Y, Zu H, Zhang J. Enhanced dissolution and stability of adefovir dipivoxil by cocrys-tal formation. J Pharm Pharmacol. 2011;63(4):483-90.

78. Kuminek G, Cao F, da Rocha, Alanny Bahia de Oliveira, Cardoso SG, Rodríguez-Hornedo N. Cocrystals to facilitate delivery of poorly soluble compounds beyond-rule-of-5. Adv Drug Deliv Rev. 2016;101:143-66.

67

79. Bethune SJ, Schultheiss N, Henck J. Improving the poor aqueous solubility of nutraceuti-cal compound pterostilbene through cocrystal formation. Cryst Growth Des. 2011;11(7):2817-23.

80. Gao P, Morozowich W. Development of supersaturatable self-emulsifying drug delivery system formulations for improving the oral absorption of poorly soluble drugs. Expert Opin Drug Deliv. 2006;3(1):97-110.

81. Guzman HR, Tawa M, Zhang Z, Ratanabanangkoon P, Shaw P, Gardner CR, et al. Com-bined use of crystalline salt forms and precipitation inhibitors to improve oral absorption of celecoxib from solid oral formulations. J Pharm Sci. 2007;96(10):2686-702.

82. Yang M, Gong W, Wang Y, Shan L, Li Y, Gao C. Bioavailability Improvement Strategies for Poorly Water-Soluble Drugs Based on the Supersaturation Mechanism: An Update. J Pharm Pharm Sci. 2016;19(2):208-25.

83. Watanabe T, Hasegawa S, Wakiyama N, Kusai A, Senna M. Prediction of apparent equi-librium solubility of indomethacin compounded with silica by 13C solid state NMR. Int J Pharm. 2002;248(1-2):123-9.

84. O'driscoll C, Griffin B. Biopharmaceutical challenges associated with drugs with low aqueous solubility—the potential impact of lipid-based formulations. Adv Drug Deliv Rev. 2008;60(6):617-24.

85. Gao P, Rush BD, Pfund WP, Huang T, Bauer JM, Morozowich W, et al. Development of a supersaturable SEDDS (S‐SEDDS) formulation of paclitaxel with improved oral bioa-vailability. J Pharm Sci. 2003;92(12):2386-98.

86. Gao P, Guyton ME, Huang T, Bauer JM, Stefanski KJ, Lu Q. Enhanced oral bioavailabil-ity of a poorly water soluble drug PNU‐91325 by supersaturatable formulations. Drug Dev Ind Pharm. 2004;30(2):221-9.

87. Simonelli A, Mehta S, Higuchi W. Inhibition of sulfathiazole crystal growth by polyvi-nylpyrrolidone. J Pharm Sci. 1970;59(5):633-8.

88. Suzuki H, Sunada H. Influence of water-soluble polymers on the dissolution of nifedipine solid dispersions with combined carriers. Chem Pharm Bull. 1998;46(3):482-7.

89. Overhoff KA, McConville JT, Yang W, Johnston KP, Peters JI, Williams RO. Effect of stabilizer on the maximum degree and extent of supersaturation and oral absorption of tacrolimus made by ultra-rapid freezing. Pharm Res. 2008;25(1):167-75.

90. Rodríguez‐hornedo N, Murphy D. Significance of controlling crystallization mechanisms and kinetics in pharmaceutical systems. J Pharm Sci. 1999;88(7):651-60.

91. Raghavan S, Trividic A, Davis A, Hadgraft J. Crystallization of hydrocortisone acetate: influence of polymers. Int J Pharm. 2001;212(2):213-21.

92. Strickley RG. Solubilizing excipients in oral and injectable formulations. Pharm Res. 2004;21(2):201-30.

93. Gao P, Akrami A, Alvarez F, Hu J, Li L, Ma C, et al. Characterization and optimization of AMG 517 supersaturatable self‐emulsifying drug delivery system (S‐SEDDS) for im-proved oral absorption. J Pharm Sci. 2009;98(2):516-28.

94. Bevernage J, Brouwers J, Brewster ME, Augustijns P. Evaluation of gastrointestinal drug supersaturation and precipitation: strategies and issues. Int J Pharm. 2013;453(1):25-35.

95. Carlert S, Pålsson A, Hanisch G, von Corswant C, Nilsson C, Lindfors L, et al. Predicting Intestinal Precipitation—A Case Example for a Basic BCS Class II Drug. Pharm Res. 2010;27(10):2119-30.

96. DiNunzio JC, Miller DA, Yang W, McGinity JW, Williams III RO. Amorphous composi-tions using concentration enhancing polymers for improved bioavailability of itracona-zole. Mol Pharm. 2008;5(6):968-80.

97. Mellaerts R, Mols R, Kayaert P, Annaert P, Van Humbeeck J, Van den Mooter G, et al. Ordered mesoporous silica induces pH-independent supersaturation of the basic low sol-

68

ubility compound itraconazole resulting in enhanced transepithelial transport. Int J Pharm. 2008;357(1-2):169-79.

98. Gu C, Rao D, Gandhi RB, Hilden J, Raghavan K. Using a novel multicompartment disso-lution system to predict the effect of gastric pH on the oral absorption of weak bases with poor intrinsic solubility. J Pharm Sci. 2005;94(1):199-208.

99. Bevernage J, Forier T, Brouwers J, Tack J, Annaert P, Augustijns P. Excipient-mediated supersaturation stabilization in human intestinal fluids. Mol Pharm. 2011;8(2):564-70.

100. Vandecruys R, Peeters J, Verreck G, Brewster ME. Use of a screening method to deter-mine excipients which optimize the extent and stability of supersaturated drug solutions and application of this system to solid formulation design. Int J Pharm. 2007;342(1-2):168-75.

101. Yamashita T, Kokubo T, Zhao C, Ohki Y. Antiprecipitant screening system for basic model compounds using bio-relevant media. Journal of the Association for Laboratory Automation. 2010;15(4):306-12.

102. Yamashita T, Ozaki S, Kushida I. Solvent shift method for anti-precipitant screening of poorly soluble drugs using biorelevant medium and dimethyl sulfoxide. Int J Pharm. 2011;419(1):170-4.

103. Warren DB, Benameur H, Porter CJ, Pouton CW. Using polymeric precipitation inhibi-tors to improve the absorption of poorly water-soluble drugs: A mechanistic basis for utility. J Drug Target. 2010;18(10):704-31.

104. Kostewicz ES, Wunderlich M, Brauns U, Becker R, Bock T, Dressman JB. Predicting the precipitation of poorly soluble weak bases upon entry in the small intestine. J Pharm Pharmacol. 2004;56(1):43-51.

105. Yang Z, Nollenberger K, Albers J, Qi S. Molecular implications of drug–polymer solu-bility in understanding the destabilization of solid dispersions by milling. Mol Pharm. 2014;11(7):2453-65.

106. Sarode AL, Malekar SA, Cote C, Worthen DR. Hydroxypropyl cellulose stabilizes amorphous solid dispersions of the poorly water soluble drug felodipine. Carbohydr Polym. 2014;112:512-9.

107. Sarode AL, Sandhu H, Shah N, Malick W, Zia H. Hot melt extrusion for amorphous solid dispersions: temperature and moisture activated drug–polymer interactions for en-hanced stability. Mol Pharm. 2013;10(10):3665-75.

108. Dalsin MC, Tale S, Reineke TM. Solution-state polymer assemblies influence BCS class II drug dissolution and supersaturation maintenance. Biomacromolecules. 2014;15(2):500-11.

109. Cheow WS, Kiew TY, Yang Y, Hadinoto K. Amorphization strategy affects the stability and supersaturation profile of amorphous drug nanoparticles. Mol Pharm. 2014;11(5):1611-20.

110. Sun DD, Lee PI. Evolution of supersaturation of amorphous pharmaceuticals: the effect of rate of supersaturation generation. Mol Pharm. 2013;10(11):4330-46.

111. Verma S, Rudraraju VS. Wetting kinetics: an alternative approach towards understand-ing the enhanced dissolution rate for amorphous solid dispersion of a poorly soluble drug. AAPS PharmSciTech. 2015;16(5):1079-90.

112. Ouellet D, Grossmann KF, Limentani G, Nebot N, Lan K, Knowles L, et al. Effects of particle size, food, and capsule shell composition on the oral bioavailability of dabraf-enib, a BRAF inhibitor, in patients with BRAF mutation-positive tumors. J Pharm Sci. 2013;102(9):3100-9.

113. Curatolo W, Nightingale JA, Herbig SM. Utility of hydroxypropylmethylcellulose ace-tate succinate (HPMC-AS) for initiation and maintenance of drug supersaturation in the GI milieu. Pharm Res. 2009;26(6):1419-31.

69

114. Prasad D, Chauhan H, Atef E. Amorphous stabilization and dissolution enhancement of amorphous ternary solid dispersions: combination of polymers showing drug–polymer interaction for synergistic effects. J Pharm Sci. 2014;103(11):3511-23.

115. Kim MS, Baek IH. Fabrication and evaluation of valsartan-polymer- surfactant compo-site nanoparticles by using the supercritical antisolvent process. Int J Nanomedicine. 2014 Nov 7;9:5167-76.

116. Jantratid E, Janssen N, Reppas C, Dressman JB. Dissolution media simulating conditions in the proximal human gastrointestinal tract: an update. Pharm Res. 2008;25(7):1663.

117. Dressman JB, Amidon GL, Reppas C, Shah VP. Dissolution testing as a prognostic tool for oral drug absorption: immediate release dosage forms. Pharm Res. 1998;15(1):11-22.

118. Bevernage J, Brouwers J, Clarysse S, Vertzoni M, Tack J, Annaert P, et al. Drug super-saturation in simulated and human intestinal fluids representing different nutritional states. J Pharm Sci. 2010;99(11):4525-34.

119. Bevernage J, Hens B, Brouwers J, Tack J, Annaert P, Augustijns P. Supersaturation in human gastric fluids. Eur J Pharm Biopharm. 2012;81(1):184-9.

120. Alonzo DE, Zhang GG, Zhou D, Gao Y, Taylor LS. Understanding the behavior of amorphous pharmaceutical systems during dissolution. Pharm Res. 2010;27(4):608-18.

121. McAllister M. Dynamic dissolution: a step closer to predictive dissolution testing? Mol Pharm. 2010;7(5):1374-87.

122. Bevernage J, Brouwers J, Annaert P, Augustijns P. Drug precipitation–permeation inter-play: supersaturation in an absorptive environment. Eur J Pharm Biopharm. 2012;82(2):424-8.

123. Kataoka M, Sugano K, da Costa Mathews C, Wong JW, Jones KL, Masaoka Y, et al. Application of dissolution/permeation system for evaluation of formulation effect on oral absorption of poorly water-soluble drugs in drug development. Pharm Res. 2012;29(6):1485-94.

124. Baldyga J, Orciuch W. Some hydrodynamic aspects of precipitation. Powder Technol. 2001;121(1):9-19.

125. Manth T, Mignon D, Offermann H. Experimental investigation of precipitation reactions under homogeneous mixing conditions. Chemical engineering science. 1996;51(11):2571-6.

126. Lindfors L, Forssén S, Westergren J, Olsson U. Nucleation and crystal growth in super-saturated solutions of a model drug. J Colloid Interface Sci. 2008;325(2):404-13.

127. Psachoulias D, Vertzoni M, Butler J, Busby D, Symillides M, Dressman J, et al. An in vitro methodology for forecasting luminal concentrations and precipitation of highly permeable lipophilic weak bases in the fasted upper small intestine. Pharm Res. 2012;29(12):3486-98.

128. D’Arcy DM, Healy AM, Corrigan OI. Towards determining appropriate hydrodynamic conditions for in vitro in vivo correlations using computational fluid dynamics. Eur J Pharm Sci. 2009;37(3-4):291-9.

129. Radebaugh GW, Murtha JL, Julian TN, Bondi JN. Methods for evaluating the puncture and shear properties of pharmaceutical polymeric films. Int J Pharm. 1988;45(1-2):39-46.

130. Garsuch V, Breitkreutz J. Comparative investigations on different polymers for the preparation of fast‐dissolving oral films. J Pharm Pharmacol. 2010;62(4):539-45.

131. Wlodarski K, Sawicki W, Paluch KJ, Tajber L, Grembecka M, Hawelek L, et al. The influence of amorphization methods on the apparent solubility and dissolution rate of tadalafil. Eur J Pharm Sci. 2014 1 October 2014;62:132-40.

132. Hobbs D, Karagianis J, Treuer T, Raskin J. An in vitro analysis of disintegration times of different formulations of olanzapine orodispersible tablet: A preliminary report. Drugs R. 2013;13(4):281-8.

70

133. Gulsun T, Gursoy RN, Oner L. Design and characterization of nanocrystal formulationscontaining ezetimibe. Chem Pharm Bull . 2011;59(1):41-5.

134. Parmar KR, Shah SR, Sheth NR. Studies in dissolution enhancement of ezetimibe bysolid dispersions in combination with a surface adsorbent. Dissolut Technol. 2011;8:55-61.

135. Taupitz T, Dressman JB, Klein S. New formulation approaches to improve solubility anddrug release from fixed dose combinations: Case examples pioglitazone/glimepiride andezetimibe/simvastatin. Eur J Pharm Biopharm. 2013 5;84(1):208-18.

136. Patterson JE, James MB, Forster AH, Lancaster RW, Butler JM, Rades T. Preparation ofglass solutions of three poorly water soluble drugs by spray drying, melt extrusion andball milling. Int J Pharm. 2007;336(1):22-34.

137. Zhang Y, Luo R, Chen Y, Ke X, Hu D, Han M. Application of Carrier and Plasticizer toImprove the Dissolution and Bioavailability of Poorly Water-Soluble Baicalein by HotMelt Extrusion. AAPS PharmSciTech. 2014;15(3):560-8.

138. Bevernage J, Hens B, Brouwers J, Tack J, Annaert P, Augustijns P. Supersaturation inhuman gastric fluids. Eur J Pharm Biopharm. 2012;81(1):184-9.

139. Laties AM. Vision disorders and phosphodiesterase type 5 inhibitors. Drug safety.2009;32(1):1-18.

140. Benet LZ, Broccatelli F, Oprea TI. BDDCS applied to over 900 drugs. The AAPS jour-nal. 2011;13(4):519-47.

141. Gresser U, Gleiter C. Erectile dysfunction: comparison of efficacy and side effects of thePDE-5 inhibitors sildenafil, vardenafil and tadalafil-review of the literature. Eur J MedRes. 2002;7(10):435-46.

142. Krupa A, Majda D, Mozgawa W, Szlęk J, Jachowicz R. Physicochemical Properties ofBosentan and Selected PDE-5 Inhibitors in the Design of Drugs for Rare Diseases. AapsPharmscitech. 2017;18(4):1318-31.

143. Alhalaweh A, Roy L, Rodríguez-Hornedo N, Velaga SP. pH-dependent solubility ofindomethacin–saccharin and carbamazepine–saccharin cocrystals in aqueous media. Mo-lecular pharmaceutics. 2012;9(9):2605-12.

144. Babu NJ, Nangia A. Solubility advantage of amorphous drugs and pharmaceutical co-crystals. Cryst Growth Des. 2011;11(7):2662-79.

145. Basavoju S, Boström D, Velaga SP. Pharmaceutical cocrystal and salts of norfloxacin.Cryst Growth Des. 2006;6(12):2699-708.

146. Chapple CR, Montorsi F, Tammela TL, Wirth M, Koldewijn E, Fernández EF, et al.Silodosin therapy for lower urinary tract symptoms in men with suspected benign pros-tatic hyperplasia: results of an international, randomized, double-blind, placebo-and ac-tive-controlled clinical trial performed in Europe. Eur Urol. 2011; 59(3):342-52.

147. Montorsi F. Profile of silodosin. European Urology Supplements. 2010;9(4):491-5.148. Alonzo DE, Gao Y, Zhou D, Mo H, Zhang GG, Taylor LS. Dissolution and precipitation

behavior of amorphous solid dispersions. J Pharm Sci. 2011;100(8):3316-31.149. Tanno F, Nishiyama Y, Kokubo H, Obara S. Evaluation of hypromellose acetate succin-

ate (HPMC-AS) as a carrier in solid dispersions. Drug Dev Ind Pharm. 2004;30(1):9-17.150. Miller-Chou BA, Koenig JL. A review of polymer dissolution. Prog Polym Sci.

2003;28(8):1223-70.151. Vuddanda PR, Montenegro-Nicolini M, Morales JO, Velaga S. Effect of plasticizers on

the physico-mechanical properties of pullulan based pharmaceutical oral films. Eur JPharm Sci. 2017;96:290-8.

152. Singh H, Narang JK, Singla YP, Narang RS, Mishra V. TPGS stabilized sublingual filmsof frovatriptan for the management of menstrual migraine: Formulation, design and anti-oxidant activity. J Drug Deliv Sci Technol. 2017;41:144-56.

71

153. Alhayali A, Tavellin S, Velaga S. Dissolution and precipitation behavior of ternary soliddispersions of ezetimibe in biorelevant media. Drug Dev Ind Pharm. 2016:1-10.

154. Guo Y, Luo J, Tan S, Otieno BO, Zhang Z. The applications of Vitamin E TPGS in drugdelivery. Eur J Pharm Sci. 2013;49(2):175-86.

155. Vuddanda PR, Montenegro-Nicolini M, Morales JO, Velaga S. Effect of surfactants anddrug load on physico-mechanical and dissolution properties of nanocrystalline tadalafil-loaded oral films. Eur J Pharm Sci. 2017;109:372-80.

156. Garsuch V, Breitkreutz J. Comparative investigations on different polymers for thepreparation of fast‐dissolving oral films. J Pharm Pharmacol. 2010;62(4):539-45.

157. Li N, Taylor LS. Tailoring supersaturation from amorphous solid dispersions. J Con-trolled Release. 2018.

158. Yang M, Gong W, Wang Y, Shan L, Li Y, Gao C. Bioavailability Improvement Strate-gies for Poorly Water-Soluble Drugs Based on the Supersaturation Mechanism: An Up-date. J Pharm Pharm Sci. 2016;19(2):208-25.

159. Psachoulias D, Vertzoni M, Goumas K, Kalioras V, Beato S, Butler J, et al. Precipitationin and supersaturation of contents of the upper small intestine after administration of twoweak bases to fasted adults. Pharm Res. 2011;28(12):3145-58.

160. Konno H, Handa T, Alonzo DE, Taylor LS. Effect of polymer type on the dissolutionprofile of amorphous solid dispersions containing felodipine. Eur J Pharm Biopharm.2008;70(2):493-9.

161. Usui F, Maeda K, Kusai A, Nishimura K, Yamamoto K. Inhibitory effects of water-soluble polymers on precipitation of RS-8359. Int J Pharm. 1997;154(1):59-66.

162. Bevernage J, Brouwers J, Brewster ME, Augustijns P. Evaluation of gastrointestinal drugsupersaturation and precipitation: strategies and issues. Int J Pharm. 2013;453(1):25-35.

163. Greco K, Bogner R. Crystallization of amorphous indomethacin during dissolution:effect of processing and annealing. Mol Pharm. 2010;7(5):1406-18.

164. Bevernage J, Brouwers J, Annaert P, Augustijns P. Drug precipitation–permeation inter-play: supersaturation in an absorptive environment. Eur J Pharm Biopharm.2012;82(2):424-8.

165. Kanzer J, Tho I, Flaten GE, Mägerlein M, Hölig P, Fricker G, et al. In‐vitro permeabilityscreening of melt extrudate formulations containing poorly water‐soluble drug com-pounds using the phospholipid vesicle‐based barrier. J Pharm Pharmacol.2010;62(11):1591-8.

166. Frank KJ, Rosenblatt KM, Westedt U, Hölig P, Rosenberg J, Mägerlein M, et al. Amor-phous solid dispersion enhances permeation of poorly soluble ABT-102: true supersatu-ration vs. apparent solubility enhancement. Int J Pharm. 2012;437(1):288-93.

167. Oswald S, Koll C, Siegmund W. Disposition of the cholesterol absorption inhibitorezetimibe in mdr1a/b (−/−) mice. J Pharm Sci. 2007;96(12):3478-84.

168. Fischer SM, Brandl M, Fricker G. Effect of the non-ionic surfactant Poloxamer 188 onpassive permeability of poorly soluble drugs across Caco-2 cell monolayers. Eur J PharmBiopharm. 2011;79(2):416-22.

169. Patel DD, Anderson BD. Adsorption of Polyvinylpyrrolidone and its Impact on Mainte-nance of Aqueous Supersaturation of Indomethacin via Crystal Growth Inhibition. JPharm Sci. 2015;104(9):2923-33.

170. Saha P, Kou JH. Effect of solubilizing excipients on permeation of poorly water-solublecompounds across Caco-2 cell monolayers. Eur J Pharm Biopharm. 2000;50(3):403-11.

171. Christfort JF, Plum J, Madsen CM, Nielsen LH, Sandau M, Andersen K, et al. Develop-ment of a Video-Microscopic Tool T167o Evaluate the Precipitation Kinetics of PoorlyWater Soluble Drugs: A Case Study with Tadalafil and HPMC. Mol Pharm. 2017;14(12):4154-60.

72

DISSERTATIONS FROM THE DEPARTMENT OF HEALTH SCIENCE, LULEÅ UNIVERSITY OF TECHNOLOGY, SWEDEN

Doctoral theses

Terttu Häggström. Life-story perspective on caring within cultural contexts: experiences of severe illness and of caring. (Nursing) 2004.

Inger Jacobson. Injuries among female football players. (Physiotherapy) 2006. Karl Elling Ellingsen. Lovregulert tvang og refleksiv praksis. (Health Science and Human Services)

2006. Annika Näslund. Dynamic ankle-foot orthoses in children with spastic diplegia: interview and

experimental studies. (Physiotherapy) 2007. Inger Lindberg. Postpartum care in transition: parents’ and midwives’ expectations and experience of

postpartum care including the use of videoconferencing. (Nursing) 2007. Åsa Widman. Det är så mycket som kan spela in – en studie av vägar till, genom och från

sjukskrivning baserad på intervjuer med långtidssjukskrivna. (Health Science and Human Services) 2007.

Eija Jumisko. Striving to become familiar with life with traumatic brain injury: experiences of people with traumatic brain injury and their close relatives. (Nursing) 2007.

Gunilla Isaksson. Det sociala nätverkets betydelse för delaktighet i dagliga aktiviteter: erfarenheter från kvinnor med ryggmärgsskada och deras män. (Health Science and Human Services) 2007.

Nina Lindelöf. Effects and experiences of high-intensity functional exercise programmes among older people with physical or cognitive impairment. (Physiotherapy) 2008.

Åsa Engström. A wish to be near: experiences of close relatives within intensive care from the perspective of close relatives, formerly critically ill people and critical care nurses. (Nursing) 2008. Catrine Kostenius. Giving voice and space to children in health promotion. (Health Science and

Human Services) 2008. Anita Melander Wikman. Ageing well: mobile ICT as a tool for empowerment of elderly people in

home health care and rehabilitation. (Physiotherapy) 2008. Sedigheh Iranmanesh. Caring for dying and meeting death: the views of Iranian and Swedish nurses

and student nurses. (Nursing) 2009. Birgitta Lindberg. When the baby is premature. Experiences of parenthood and getting support via

videoconferencing. (Nursing) 2009. Malin Olsson. Meaning of women’s experiences of living with multiple sclerosis. (Nursing) 2010. Lars Jacobsson. Long-term outcome after traumatic brain injury. Studies of individuals from northern

Sweden. (Health Science) 2010. Irene Wikman. Fall, perceived fall risk and activity curtailment among older people receiving home-

help services. (Physiotherapy) 2011. Christina Harrefors. God vård och användning av digitala hjälpmedel. Föreställningar hos äldre och

vårdpersonal. (Nursing) 2011. Agneta Larsson. Identifying, describing and promoting health and work ability in a workplace

context. (Physiotherapy) 2011. Lisbeth Eriksson. Telerehabilitation: Physiotherapy at a distance at home. (Physiotherapy) 2011. Amjad Alhalaweh. Pharmaceutical Cocrystals: Formation mechanisms, solubility behaviour and solid-

state properties. (Health Science) 2012. Katarina Mikaelsson. Fysisk aktivitet, inaktivitet och kapacitet hos gymnasieungdomar.

(Physiotherapy) 2012.

Carina Nilsson. Information and communication technology as a tool for support in home care. - Experiences of middle-aged people with serious chronic illness and district nurses. (Nursing) 2012.

Britt-Marie Wällivaara. Contemporary home-based care: encounters, relationships and the use of distance-spanning technology. (Nursing) 2012.

Stina Rutberg. Striving for control and acceptance to feel well. Experiences of living with migraine and attending physical therapy. (Physiotherapy) 2013.

Päivi Juuso. Meanings of women's experiences of living with fibromyalgia. (Nursing) 2013. Anneli Nyman. Togetherness in Everyday Occupations. How Participation in On-Going Life with

Others Enables Change. (Occupational therapy) 2013. Caroline Stridsman. Living with chronic obstructive pulmonary disease with focus on fatigue, health

and well-being. (Nursing) 2013. Ann-Sofie Forslund. A Second Chance at Life: A Study About People Suffering Out-Of-Hospital

Cardiac Arrest. (Nursing) 2014. Birgitta Nordström. Experiences of standing in standing devices: voices from adults, children and

their parents. (Physiotherapy) 2014. Malin Mattsson. Patients’ experiences and patient-reported outcome measures in systemic lupus

erythematosus and systemic sclerosis. (Physiotherapy) 2014. Eva Lindgren. “It’s all about survival”: Young adults’ transitions within psychiatric care from the

perspective of young adults, relatives, and professionals. (Nursing) 2014. Annette Johansson. Implementation of Videoconsultation to Increase Accessibility to Care and

Specialist Care in Rural Areas: - Residents, patients and healthcare personnel´s views. (Nursing) 2015.

Cecilia Björklund. Temporal patterns of daily occupations and personal projects relevant for older persons’ subjective health: a health promotive perspective. (Occupational therapy) 2015.

Ann-Charlotte Kassberg. Förmåga att använda vardagsteknik efter förvärvad hjärnskada: med fokus mot arbete. (Occupational therapy) 2015.

Angelica Forsberg. Patients' experiences of undergoing surgery: From vulnerability towards recovery -including a new, altered life. (Nursing) 2015.

Maria Andersson Marchesoni. “Just deal with it” Health and social care staff´s perspectives on changing work routines by introducing ICT: Perspectives on the process and interpretation of values. (Nursing) 2015.

Sari-Anne Wiklund-Axelsson. Prerequisites for sustainable life style changes among older persons with obesity and for ICT support. (Physiotherapy) 2015.

Anna-Karin Lindqvist. Promoting adolescents' physical activity @ school. (Physiotherapy) 2015. Ulrica Lundström. Everday life while aging with a traumatic spinal cord injury. (Occupational

therapy) 2015. Sebastian Gabrielsson. A moral endeavour in a demoralizing context: Psychiatric inpatient care from

the perspective of professional caregivers. (Nursing) 2015. Git-Marie Ejneborn-Looi. Omvårdnad som reflekterande praktik: Att se och använda alternativ till

tvång i psykiatrisk vård. (Nursing) 2015. Sofi Nordmark. Hindrances and Feasibilities that Affect Discharge Planning: Perspectives Before and

After the Development and Testing of ICT Solutions. (Nursing) 2016. Marianne Sirkka. Hållbart förbättringsarbete med fokus på arbetsterapi och team Möjligheter och

utmaningar. (Occupational therapy) 2016. Catharina Nordin. Patientdelaktighet och behandlingseffekter inom multimodal rehabilitering och

web-baserat beteendeförändringsprogram för aktivitet. (Physiotherapy) 2016. Silje Gustafsson. Self-care for Minor Illness: People's Experiences and Needs. (Nursing) 2016. Anna Nygren Zotterman. Encounters in primary healthcare from the perspectives of people with

long-term illness, their close relatives and district nurses. (Nursing) 2017.

Katarina Leijon Sundqvist. Evaluation of hand skin temperature -Infrared thermography in combination with cold stress tests. (Health Science) 2017.

Tommy Calner. Persistent musculoskeletal pain: A web-based activity programme for behaviour change, does it work? Expectations and experiences of the physiotherapy treatment process. (Physiotherapy) 2016.

Licentiate theses

Marja Öhman. Living with serious chronic illness from the perspective ofpeople with serious chronic illness, close relatives and district nurses. (Nursing) 2003.

Kerstin Nyström. Experiences of parenthood and parental support during the child's first year. (Nursing) 2004.

Eija Jumisko. Being forced to live a different everyday life: the experiences of people with traumatic brain injury and those of their close relatives. (Nursing) 2005.

Åsa Engström. Close relatives of critically ill persons in intensive and critical care: the experiences of close relatives and critical care nurses. (Nursing) 2006.

Anita Melander Wikman. Empowerment in living practice: mobile ICT as a tool for empowerment of elderly people in home health care. (Physiotherapy) 2007.

Carina Nilsson. Using information and communication technology to support people with serious chronic illness living at home. (Nursing) 2007.

Malin Olsson. Expressions of freedom in everyday life: the meaning of women's experiences of living with multiple sclerosis. (Nursing) 2007.

Lena Widerlund. Nya perspektiv men inarbetad praxis: en studie av utvecklingsstördas delaktighet och självbestämmande. (Health Science and Human Services) 2007.

Birgitta Lindberg. Fathers’ experiences of having an infant born prematurely. (Nursing) 2007. Christina Harrefors. Elderly people’s perception about care and the use of assistive technology

services (ATS). (Nursing) 2009. Lisbeth Eriksson. Effects and patients' experiences of interactive video-based physiotherapy at home

after shoulder joint replacement. (Physiotherapy) 2009. Britt-Marie Wälivaara. Mobile distance-spanning technology in home care. Views and reasoning

among persons in need of health care and general practitioners. (Nursing) 2009. Anita Lindén. Vardagsteknik. Hinder och möjligheter efter förvärvad hjärnskada. (Health Science)

2009. Ann-Louise Lövgren Engström. Användning av vardagsteknik i dagliga aktiviteter - svårigheter och

strategier hos personer med förvärvad hjärnskada. (Health Science) 2010. Malin Mattsson. Frågeformulär och patientupplevelser vid systemisk lupus erythematosus

-en metodstudie och en kvalitativ studie. (Physiotherapy) 2011. Catharina Nordin. Patients’ experiences of patient participation prior to and within multimodal pain

rehabilitation. (Physiotherapy) 2013. Hamzah Ahmed. Relationship Between Crystal Structure and Mechanical Properties in Cocrystals

and Salts of Paracetamol. (Health Science) 2014. Johan Kruse. Experiences of Interns and General Practitioners of Communicating with and Writing

Referrals to the Radiology Department. (Health Science) 2016. For purchase information: Department of Health Science, Luleå University of Technology, S-971 87 Luleå, Sweden.

Paper I

RESEARCH ARTICLE

Dissolution and precipitation behavior of ternary solid dispersions of ezetimibe inbiorelevant media

Amani Alhayalia, Staffan Tavellinb and Sitaram Velagaa

aDepartment of Health Sciences, Division of Medical Sciences, Luleå University of Technology, Lulea, Sweden; bDepartments of Pharmacologyand Clinical Neuroscience, Umea University, Umeå, Sweden

ABSTRACTThe effects of different formulations and processes on inducing and maintaining the supersaturation ofternary solid dispersions of ezetimibe (EZ) in two biorelevant media fasted-state simulated intestinal fluid(FaSSIF) and fasted-state simulated gastric fluid (FaSSGF) at different temperatures (25 �C and 37 �C) wereinvestigated in this work.Ternary solid dispersions of EZ were prepared by adding polymer PVP-K30 and surfactant poloxamer 188using melt-quenching and spray-drying methods. The resulting solid dispersions were characterized usingscanning electron microscopy, differential scanning calorimetry (DSC), modulated DSC, powder X-ray dif-fraction and Fourier transformation infrared spectroscopy. The dissolution of all the ternary solid disper-sions was tested in vitro under non-sink conditions.All the prepared solid dispersions were amorphous in nature. In FaSSIF at 25 �C, the melt-quenched (MQ)solid dispersions of EZ were more soluble than the spray-dried (SD) solid dispersions and supersaturationwas maintained. However, at 37 �C, rapid and variable precipitation behavior was observed for all the MQand SD formulations. In FaSSGF, the melting method resulted in better solubility than the spray-dryingmethod at both temperatures.Ternary solid dispersions show potential for improving solubility and supersaturation. However, powderdissolution experiments of these solid dispersions of EZ at 25 �C may not predict the supersaturationbehavior at physiologically relevant temperatures.

ARTICLE HISTORYReceived 4 February 2016Revised 19 June 2016Accepted 6 July 2016Published online 25 August2016

KEYWORDSPoorly soluble drugs;ternary solid dispersions;supersaturation; biorelevantmedia; dissolution

Introduction

Supersaturation is becoming an important tool for increasing bio-availability as the number of poorly water soluble compoundsincreases in drug discovery processes1. Supersaturation of an orallydelivered drug is reached when its intraluminal concentrationexceeds its thermodynamic solubility equilibrium concentration2.The need for designing formulations that can create this highintestinal drug concentration is gaining more attention, becausethe induction of supersaturation in the intestinal lumen willimprove intestinal absorption as well as the therapeutic effect1.

Solid dispersion technology is an interesting option for achiev-ing supersaturated drug concentrations. The drug is dispersedwithin a polymer at the molecular level to form an amorphoussolid dispersion or solid solution3–5. Methods that have been usedto prepare solid dispersions include micronization, melt methods,spray drying and freeze drying, among others6,7. A wide variety ofpharmaceutical excipients (including polymers4,8–10 and surfac-tants10,11) have been used to stabilize the drug in the solid, super-saturated state in vivo.

In ternary solid dispersions, which have been shown toimprove the solubility12,13 and supersaturation solubility5 of poorlywater soluble and insoluble active pharmaceutical ingredients(APIs), the API is dispersed within a carrier system containing apolymer and a surfactant. The polymer is added to stabilize thesystem both in the solid state and in solution by preventing

precipitation of the drug and phase separation. The surfactantenhances the solubility and dissolution of the API by loweringthe surface tension of the drug particles and maintainingsupersaturation during dissolution12–14. Surfactants can also act asplasticizers that aid and facilitate thermal processing of the poly-mer and API15.

Ezetimibe (EZ), a cholesterol-lowering drug used in the treat-ment of atherosclerosis, was chosen as the model drug in thisstudy. It acts by inhibiting the uptake of dietary and biliary choles-terol in the intestine without affecting the uptake of fat-solublevitamins, triglycerides or bile acids. EZ is a lipophilic molecule witha log P (octanol/water) of 4.5. It is very weakly acidic (close to neu-tral) with a pKa of 10.2 (aromatic hydroxyl group)16 and thus ispractically insoluble in water, with a consequently low dissolutionrate in the gastric lumen, potentially leading to dissolution-limitedbioavailability17,18. In fact, EZ has been classified as a class IIdrug19. A number of formulations and solid-state strategies havebeen used to improve its solubility and dissolution, including bin-ary solid dispersions, nanocrystals and cocrystals16,17,19.

In this study, ternary solid dispersions of EZ were preparedusing different proportions of the polymer polyvinylpyrrolidone(PVP)-K30 and the surfactant poloxamer 188. Two preparationmethods were employed: melt quenching and spray drying.The apparent solubility of EZ from the prepared ternary solid dis-persions was studied in two simulated gastrointestinal fluids

CONTACT Sitaram Velaga sitram.velaga@ltu.se Department of Health Sciences, Division of Medical Sciences, Luleå University of Technology, Lulea 971 87,Sweden

Supplemental data for this article can be accessed here.

� 2016 Informa UK Limited, trading as Taylor & Francis Group

DRUG DEVELOPMENT AND INDUSTRIAL PHARMACY, 2016http://dx.doi.org/10.1080/03639045.2016.1220566

[fasting-state simulated intestinal fluid (FaSSIF; pH 6.5) and fasting-state simulated gastric fluid (FaSSGF; pH 1.6)] in terms of supersat-uration at two temperatures: 25 �C and 37 �C.

The overall objective of the work was to improve under-standing of supersaturation phenomena in ternary solid disper-sions of EZ in biorelevant media and to investigate the effectsof factors such as the preparation process and the temperatureof the dissolution medium on the maintenance of supersatur-ation. The specific aims of the study were (a) to prepare andcharacterize ternary solid dispersions of EZ using melt-quenchingand spray-drying methods, (b) to investigate the powder dissol-ution and precipitation behavior (supersaturation) of ternarysolid dispersions in vitro in simulated gastric (FaSSGF) and intes-tinal (FaSSIF) fluids at 25 �C and 37 �C, and (c) to gain insightinto the effects of excipients, preparation processes, dissolutionmedia and temperature on the extent and duration of supersat-uration in vitro.

Materials and methods

Materials

Ezetimibe and PVP-K30 were purchased from Sigma-Aldrich(Stockholm, Sweden). Poloxamer 188 was a gift from BASFChemicals (Ludwigshafen, Germany). Simulated intestinal fluidpowders for preparing FaSSIF (sodium taurocholate, lecithin,sodium chloride, sodium hydroxide and monobasic sodium phos-phate) and FaSSGF (sodium taurocholate, lecithin, sodium chlorideand hydrochloric acid) were purchased from biorelevant.com(London, UK). High performance liquid chromatography (HPLC)grade acetonitrile was purchased from Merck Millipore (Solna,Sweden). All other reagents and solvents were of analytical grade.PTFT syringe filters 0.2mm were used for all the experiments. Verypure deionized water (Millipore, Solna, Sweden) was used through-out the study. All chemicals and solvents were used without fur-ther purification.

Preparation of solid dispersions

Melt-quenched (MQ) solid dispersionsPrior to the preparation of solid dispersions, EZ and PVP-K30 werekept in the oven at 100 �C for 24 h to remove any adsorbed waterfrom the solids, and differential scanning calorimetry (DSC) ana-lysis was performed on the dried crystalline EZ to ensure its crys-talline nature (data not shown). Crystalline EZ, PVP-K30 andpoloxamer 188 in different proportions (see Table 1) were accur-ately weighed and mixed gently for 1–2min using a porcelainmortar and pestle. The resulting powder mixtures were placed ina preheated oven (BINDER, Tuttlingen, Germany) at 200 �C for5–8min. After complete melting, the liquid was immediately putin a freezer at �20 �C for 1–2 days. The MQ samples were thenstored in a desiccator over silica until the day of analysis. Beforethe analysis, all the formulations were ground and sieved using125 mm sieves.

Using a similar method, the amorphous form of EZ was pre-pared for comparison.

Spray-dried (SD) solid dispersionsCrystalline EZ, PVP-K30 and poloxamer 188 in different proportionswere accurately weighed and dissolved (equivalent to 2% w/v ofsolids) in methanol. The solutions were spray dried using a Buchimini spray dryer B-290 attached to an inert loop B-295 to trap theresidual solvents. The processing conditions were as follows: inlettemperature 80 �C, air flow 357 L h�1, aspiration 70–80%, feed-flowrate 3mL/min and outlet temperature 40–50 �C. Nitrogen gas wasused as the drying gas in a closed-loop experiment. The collectedSD powders were stored in a sealed desiccator over silica until theday of analysis.

Scanning electron microscopy (SEM)The morphology of the particles was examined using a Merlinscanning electron microscope (Zeiss, Oberkochen, Germany)equipped with X-Max 50mm2 X-ray detectors (Oxford Instruments,Abingdon, UK). All the samples were coated with tungsten toincrease the conductivity of the electron beam. The instrumentvoltage was 15 kV and the current was 1 nA.

Solid-state and powder characterization

Thermo-gravimetric analysis (TGA)The weight loss and thermal degradation behavior of all the for-mulation components (EZ, PVP-K30 and poloxamer 188) wereinvestigated using a Thermal Advantage TGAQ5000 (TAInstruments, New Castle, DE) connected to a cooling system.Powdered samples (4–7mg) were placed in the aluminum pansand the furnace was heated from 20 �C to 550 �C at a rate of 5 �C/min. Nitrogen at a flow rate of 50mL/min was used as a purginggas. Universal analysis software was used to analyze the results.

Differential scanning calorimetryThe thermal behavior of the raw materials and solid dispersionswas studied using a DSC Q1000 (TA Instruments, New Castle, DE).The instrument was calibrated before use for temperature andenthalpy using indium. About 1–4mg of the sample was accur-ately weighed and placed in a non-hermetic aluminum pan, whichwas then crimp-sealed. The samples were heated from 25 �C to200 �C at a heating rate of 10 �C/min under continuous nitrogenpurge (50mL/min). The calorimeter was equipped with a refriger-ated cooling system. The data were analyzed using TA analysissoftware (TA Instruments, New Castle, DE).

Modulated differential scanning calorimetry (MDSC)MDSC was performed using a Q1000 instrument (TA Instruments,New Castle, DE) for better identification of the glass transitiontemperature (Tg) through the separation of reversible (i.e. Tg)from nonreversible (i.e. enthalpy recovery, fusion) thermal events.Samples (2–5mg) were accurately weighed and placed in alumi-num nonhermetic pans, which were then crimped. These sampleswere then heated from 25 �C to 200 �C at 5 �C/min with modula-tion amplitude of ±0.80 �C every 60 s.

Powder X-ray diffraction (PXRD)PXRD patterns for the crystalline and amorphous EZ and ternarysolid dispersions were obtained using an Empyrean PXRD instru-ment (PANalytical, Almelo, The Netherlands) equipped with a Pixel3D detector and monochromatic Cu Ka radiation. The tube voltageand amperage were 45 kV and 40mA, respectively. Powdered

Table 1. Proportions of different components in the melt-quenched (MQ) andspray-dried (SD) solid dispersions.

Sample ID Ezetimibe (%w/w) PVP-K30 (%w/w) Poloxamer188 (%w/w)

MQ1 or SD1 5 90 5MQ2 or SD2 15 80 5MQ3 or SD3 10 80 10

2 A. ALHAYALI ET AL.

samples were loaded into the oval cavity in the metal sampleholder and carefully leveled. The experimental settings were as fol-lows: 2h ranged from 5� to 40�, step size 0.02� 2h. The data wereprocessed using High Score plus Version 3.0 software (PANalytical,Almelo, The Netherlands).

Fourier transformation infrared spectroscopy (FTIR)A Bruker IFS66v/S FTIR spectrometer equipped with a deuteratedtriglycine sulfate detector was used to obtain the FTIR spectra ofthe samples. Powdered samples were mixed with KBr and IR spec-tra were obtained in the diffuse reflectant mode (Kubelka Munk).The following experimental settings were used: 64 scans with aresolution of 4 cm�1, spectral region 400–4000 cm�1.

In vitro dissolution study

Excess amounts of the SD or MQ solid dispersions were suspendedin 10mL of FaSSIF and FaSSGF in sealed test tubes and placed ina preheated shaking water bath at a shaking speed of 60 rpm(orbital shaking). The experiments were performed at 25 �C and37 �C. The samples were withdrawn after 5, 15, 30, 45, 60, 75, 90,120, 180 and 240min and centrifuged (5min, 17,000�g), then fil-tered through 0.2mm PTFE syringe filters to ensure complete sep-aration of the solid phase. The filtrate was diluted adequately withmobile phase and then analyzed by HPLC.

For comparison, the dissolution properties of crystalline andamorphous EZ in FaSSIF and FaSSGF were also studied. The sam-ples were withdrawn at pre-defined intervals following the aboveprocedure. All experiments were done in triplicate.

High performance liquid chromatography

The concentration of EZ was determined in an HPLC 1290 seriesinstrument equipped with a quaternary pump and a UV detector(Agilent, Santa Clara, CA). A C-18 Eclipse plus 3 mm (4.6mm�100mm) column was used. The mobile phase consisted of 40:60(v/v) 0.05% sodium 1-heptane sulfonic acid and acetonitrile. Theisocratic method was used, with a flow rate of 0.5mL/min. Theinjection volume was 10mL and the measurements were carriedout at a wavelength of 230 nm, in triplicate.

Results and discussion

Preparation of ternary solid dispersions

Melt-quenching and spray-drying methods were used to prepareternary solid dispersions. Melting and quenching, one of the old-est methods used in the preparation of solid dispersions, relieson mixing the molten drug with a molten or supercooled poly-mer followed by rapid cooling or solidification6. With the help ofpreliminary experiments and prior knowledge, the drug load inthe formulations was varied in the range of 5–15% w/w (Table 1).The premixed EZ and excipients were then subjected to meltingand quenching. The process temperature of 200 �C was basedon the melting temperature and thermal decomposition temper-atures of the raw materials (see Figure S1). Visually glassy mate-rials of pure EZ and the solid dispersions were obtained. Spraydrying is a widely used and efficient method of producing soliddispersions. It is a solvent-based bottom-up method that relieson the rapid evaporation of a solution containing the API andpolymers, to yield fine powders20. In addition to formulationvariables, process parameters such as the inlet temperature,

feed rate, flow rate of atomization air, etc. are known to affectthe solid-state and physicochemical properties of the resultingsolid dispersions6. In this study, the formulation of EZ, polox-amer and polymer was varied under fixed spray-drying processconditions (Table 1). Dry powders of EZ and solid dispersionswere obtained.

Figure 1 shows SEM micrographs of crystalline and amorphousEZ and the solid dispersions obtained using the melt-quenchingand spray-drying methods. As can be seen, raw EZ is formed asrod/block-shaped crystals. The ground particles of the solid disper-sions prepared by melt-quenching were irregular in shape, pos-sibly as a result of micronization of the glassy material, while SDparticles were uniform, spherical, porous and aggregated.However, there were no major differences in particle morphologybetween the different formulations. Further, there were no EZcrystals on the surfaces of the particles (Figure 1). Under fixed pro-cess conditions, SD particle formation was not affected by formu-lation changes.

Solid-state characterization of materials

Thermal analysisTGA was used to determine weight loss due to degradation ofraw materials and the loss of residual solvents/moisture in theprocessed samples (MQ and SD). Figure S1 (see supplementary fig-ure) shows the TGA weight loss curves for EZ, PVP-K30 and polox-amer 188. It can be seen that EZ, PVP-K30 and poloxamer lostsignificant weight, attributed to thermal decomposition at 348 �C,401 �C and 453 �C, respectively. It is important to know the ther-mal decomposition temperatures in order to avoid the risk of theAPI or excipients degrading during the preparation of solid disper-sions by thermal methods such as melting and quenching andmelting followed by extrusion21. It was invaluable information forthe preparation of the solid dispersions in this study.

The final product in solvent-based methods like spray dryingmay contain residual solvents that can compromise the stability ofthe solid dispersion. TGA weight loss curves for selected solid dis-persions are shown in Figure 2. It is clear that the residual solventlevels are ranging between 3.8 and 5.8% in the MQ and SD soliddispersions, suggesting efficient drying of the MQ solid dispersionproduct.

DSC analysis results for crystalline EZ, amorphous EZ, PVP-K30 and poloxamer 188 are shown in Figure 3. In the DSC

Figure 1. Thermo-gravimetric analysis curves of melt-quenched (MQ1) and spray-dried (SD1) solid dispersions of EZ showing moisture content.

DRUG DEVELOPMENT AND INDUSTRIAL PHARMACY 3

thermograms for EZ, an endothermic peak corresponding to theEZ melting point (Tm¼ 163.44 �C) was observed. The Tg forPVP-K30 appeared at 160.7 �C and a broad endothermic bandwas attributed to the sorbed moisture. In fact, the hygroscopic

nature of PVP is well known22. Poloxamer 188 showed an endo-thermic transition at 54.34 �C which corresponded to the melt-ing point, confirming its crystalline nature. The thermal behaviorof these raw materials was in agreement with that in previousreports18,21. In the DSC thermogram for amorphous EZ,the change in heat capacity at 60.14 �C was followed by an exo-thermic transition at 129.34 �C and an endotherm at 163 �C(Figure 3(b)). These thermal events were attributed to the Tg,recrystallization and melting points typical of an amorphousmaterial. The DSC thermograms of these ternary solid disper-sions had overlapping events, which made interpretation of theresults difficult.

MDSC is a powerful tool that separates the total heat flow sig-nal into reversing and non-reversing parts. The reversing heat flowalways contains the Tg while the non-reversing heat flow containsthe kinetic transitions, such as crystallization, evaporation, degrad-ation, etc. MDSC was carried out on solid dispersions prepared bymelt-quenching and spray-drying methods to separate the Tgfrom the dehydration peak. Figure 4 shows the MDSC reversingheat flow thermograms for different solid dispersions. The Tg wasbetter resolved and a single Tg (no melting peak for EZ) wasobserved at 140–180 �C for all solid dispersions. This suggests thatEZ was molecularly dispersed within the polymer and surfactant,forming an amorphous solid solution23–25. The poloxomer couldhave increased the solubility of the drug in the polymer, aidingthe formation of a solid solution14,26.

Figure 2. Scanning electron microscope photographs of (a) crystalline ezetimibe (EZ), (b) amorphous EZ, and (c) melt-quenched (MQ1) and (d) spray-dried (SD1) soliddispersions of EZ.

Figure 3. Differential scanning microscopy thermograms of (a) crystalline ezeti-mibe (EZ), (b) amorphous EZ, (c) polyvinylpyrrolidone (PVP)-K30 and (d) polox-amer 188.

4 A. ALHAYALI ET AL.

Powder X-ray diffraction (PXRD)Figure 5 shows the PXRD patterns for raw and amorphous EZ andsolid dispersions prepared by spray drying. Raw EZ showed severalcharacteristic peaks at 2h angles of 8.2�, 9.8�, 13.5�, 16.3�, 19.0�,20.1�, 23.6� and 25.5�, confirming its crystalline nature. The PXRDpattern for crystalline EZ was similar to a previously published pat-tern18. The diffraction peaks completely disappeared in the PXRDpattern of amorphous EZ, resulting in the diffuse hollow patterndistinctive of amorphous materials. All SD solid dispersions(SD1–3) showed diffuse halo PXRD patterns, confirming theamorphous nature of the solid dispersion; i.e. the API was com-pletely dispersed in the polymer and the surfactant matrix15.Similar PXRD spectra were obtained for solid dispersions preparedby melt-quenching, confirming their amorphous nature (Figure 5).

Fourier transformation infrared spectroscopy (FTIR)Diffuse reflectance IR was used to investigate the possible molecu-lar interactions between the drug and polymers in the solid

Figure 4. Modulated differential scanning calorimetry reversing heat-flow thermograms of (a) melt-quenched (MQ) solid dispersion 1, (b) MQ2 and (c) MQ3; and (d)spray-dried (SD) solid dispersion 1, (e) SD2 and (f) SD3 showing the heat capacity (Tg) in the region (140� 180 �C).

Figure 5. Powder X-ray diffraction patterns of (a) crystalline ezetimibe (EZ), (b)amorphous EZ, and (c) melt-quenched (MQ1) and (d) spray-dried (SD1) solid dis-persions of EZ.

DRUG DEVELOPMENT AND INDUSTRIAL PHARMACY 5

dispersions. IR spectroscopy has been extensively used to studyinteractions between the components of solid dispersions and toaid understanding of the mechanisms behind stabilization andsupersaturation behavior27. Figure 6 depicts the IR spectra for crys-talline EZ, amorphous EZ, the physical mixtures of EZ, and selectedsolid dispersions from melt-quenching and spray-drying methods(MQ2 and SD2 were used as representative solid dispersions inFigure 6). In the IR spectrum of crystalline EZ, a number of charac-teristic vibrational bands were identified at 3440 cm�1 (O–Hstretch), 2919 cm�1 (C–H stretch), 1728 cm�1 (C¼O stretch),1603 cm�1 (C¼C stretch), 1443.9 cm�1 (C–N stretch), 1398.2 and1223.0 cm�1 (C–F stretch) and 834.8 cm�1 (ring vibration of para-disubstituted benzene) (Figure 6). The IR spectrum for crystallineEZ was in good agreement with the spectra presented in previousarticles18,28. In the IR spectrum for amorphous EZ, the vibrationalbands were broadened in general, and a number of bands hadshifted significantly (Figure 6). Clearly, the band corresponding tothe O–H stretch at 3440 cm�1 was broadened, with a peak at3321 cm�1, and the C¼O stretching band at 1728 cm�1 hadmoved to 1723 cm�1. Bathochromic shifts in carbonyl band posi-tions (i.e. lower wave numbers) are indicative of electronic polar-ization of the double bond as a result of interactions between thelone pair of electrons on the oxygen and hydrogen donor groups.These changes could be attributed to changes in the intermolecu-lar H-bonding interactions between these functional groups in thecrystalline and amorphous forms of EZ. Similar observations havebeen made with many pharmaceutical molecules including indo-methacin, celecoxib, furosemide, etc.29–32. In the IR spectrum ofPVP-K30, a broad band was observed in the range1600–1700 cm�1; this was assigned to the carbonyl stretchingvibrations (data not shown). Poloxamer 188 exhibited characteris-tic peaks at 3000–2800 and 1100 cm�1, corresponding to stretch-ing vibrations of the O–H, C–H and C–O groups (data not shown).

Though dilution effects were seen, the IR spectra of the physicalmixtures were summations of the individual components, indicat-ing no interactions between the components (Figure 6). In con-trast, in the IR spectrum of the MQ solid dispersion MQ2,significant changes were seen: the O–H stretching band in EZ atabout 3440 cm�1 had disappeared and a C¼O stretch band at1728 cm�1 had shifted to a lower wave number and broadened.Similarly, changes in the vibrational bands were observed in the IRspectra of the SD dispersions (SD2). These changes in the IR spec-tra of solid dispersions could be attributed to the formation ofstrong interactions (H-binding, van der Waals, etc.) between EZand the PVP. The hydrogen bonding interactions between OH orNH groups on the drug molecules and the carbonyl group on PVPhave been cited in many studies previously31,32. These interactionsbetween the API and the excipients are believed to prevent self-association (dimerization) and nucleation of the API phase31–33.The mixing effects on the stretching vibration modes of C¼O,C¼C, C–N and C–F in the range 1200–1730 cm�1 are more prom-inent in MQ solid dispersions than in SD systems, probably relatedto the level of miscibility between the components.

In vitro dissolution study

It is important to optimize the in vitro experimental conditions instudies of the supersaturation behavior of amorphous solids toensure that they are biorelevant to the in vivo conditions. Thisstudy used biorelevant test media, non-sink conditions and orbitalshaking. It has been suggested that non-sink conditions are morebiorelevant for evaluating the dissolution behavior of supersatura-ble formulations where the maximum solution concentration aredetermined9. The effect of hydrodynamics on the supersaturation/precipitation is poorly understood. In a recent paper, the precipita-tion rate was shown to be remarkably slower in the shaking modelthan in the stirring model, emphasizing the importance of theapplied hydrodynamic methodology34.

Effects of process, formulation and temperature onsupersaturation in FaSSIFFigure 7 shows the powder dissolution (concentration versustime) profiles for the prepared solid dispersions in FaSSIF. Thesolubility of crystalline EZ in FaSSIF (pH 6.5) at 25 �C and 37 �Cwas 5.9 mg/mL and 10.1 mg/mL, respectively (Figure 7(a,b)). Theapparent solubility (concentration in solution) of EZ from theamorphous form increased and then rapidly decreased at 25 �C(Figure 7(a,b)). In contrast, there was no supersaturated concentra-tion at 37 �C for amorphous EZ (Figure 7(c,d)). This phenomenonof supersaturation and rapid precipitation is typical of amorphoussolids and is described as the “spring” effect3. For orally adminis-tered formulations, the dose-to-solubility ratio has been suggestedas an indicator of the precipitation propensity of the drug in theintestinal lumen35. Drugs with dose-to-solubility (non-ionized)ratios of more than 50:1 have been reported to show a tendencyto precipitate in the intestinal environment36. Evidently, the dose-to-solubility ratio for EZ is about 1000 at a dose of 10mg, suggest-ing that the observed precipitation of stable phases is thermo-dynamically favored in FaSSIF. Interestingly, the “spring” effectdisappeared (i.e. there was no increase in solubility) as the tem-perature of the dissolution experiment was increased to 37 �C.This could be explained by differences in the relationship betweendissolution and the precipitation/crystallization kinetics of crystal-line EZ. It is possible that the crystallization of amorphous EZ was

Figure 6. Fourier transformation infrared spectroscopy, showing two regions(600–2000 cm�1 and 2600–3600 cm�1) of (a) crystalline ezetimibe (EZ), (b)amorphous EZ, (c) a physical mixture of EZ, and (d) melt-quenched (MQ2) and (e)spray-dried (SD2) solid dispersions of EZ.

6 A. ALHAYALI ET AL.

faster at 37 �C. Similar behavior has been observed with amorph-ous felodipine and indomethacin in another study37.

Solid dispersions where the drug and polymers are intimatelymixed are attractive and widely used formulation strategies to pre-vent the crystallization of the stable phase and maintain supersat-uration of drugs like EZ. In this study, PVP-K30 and poloxomer 188were selected from the preliminary studies as the precipitationinhibitor and solubilizer, respectively. In fact, PVP has been wellestablished as a crystal growth inhibitor for a range of amorphousdrugs38. In the MQ solid dispersions, solution concentrations of EZwere maintained at about 50–55mg/mL and no precipitation wasobserved for at least 240min at 25 �C (Figure 7(a)). Apparently, thepresence of the polymer sustained supersaturation, with solutionconcentrations about 50 times higher than the equilibrium solubil-ity of EZ, possibly by preventing precipitation of the stable phase.However, no differences in the solution concentrations wereobserved for different polymer/drug ratios in these solid disper-sions. Interestingly, in the SD solid dispersions, supersaturation(with solution concentrations about 70–80 times higher) was fol-lowed by desupersaturation at different rates (Figure 7(b)). SDsolid dispersions generated higher initial solution concentrationsthan those generated by MQ solid dispersions, which could beexplained by differences in the particle size/morphology andmolecular miscibility (Figure 2). This could also mean that soliddispersions result in a greater driving force for crystallization. Thedesupersaturation curves also show that the lower the polymer-to-drug ratio is in the solid dispersions, the faster the precipitationkinetics will be (Figure 7(b)). The presence of 10% poloxomerresulted in even faster precipitation of the crystalline phase. Therich drug–polymer interactions (confirmed by FTIR) and changes inviscosity can alter the crystal growth and surface integration kinet-ics, leading to differences in the crystallization kinetics. Previouslypublished articles have also proposed similar mechanisms toexplain the differences in the precipitation kinetics of manyamorphous drugs39.

In general, solid dispersions prepared by the melting methodshowed slightly higher initial concentrations and faster precipita-tion kinetics at 37 �C than at 25 �C (Figure 7(a,c)), while solid dis-persions prepared by the spray-drying method showed directprecipitation in FaSSIF at 37 �C but not at 25 �C (Figure 7(b,d)).The presence of precipitation inhibitors had only a minor effect inmaintaining supersaturation at 37 �C. These results are perfectly inline with previous observations on the supersaturation and pre-cipitation behavior of other amorphous drugs in the presence ofdifferent polymers40. The explanation lies in the increase in therate of dissolution (supersaturation) and precipitation kinetics athigher temperatures. The results suggest that dissolution experi-ments performed at 25 �C may not be biorelevant and may fail topredict in vivo behavior41. In the MQ solid dispersions, desupersa-turation was dependent on the formulation: higher polymer/poloxomer content resulted in a slower precipitation rate (Figure7(c)). However, in the SD solid dispersions, precipitation was rapidirrespective of the formulation or polymer content (Figure 7(d)).This suggests that the supersaturation and precipitation behaviorof amorphous solids is sensitive to the processing methods, pos-sibly because of the history of the material and the presence ofnuclei42.

Effects of process, formulation and temperature onsupersaturation in FaSSGFFigure 8 shows the powder dissolution (concentration versus time)profiles for EZ samples in FaSSGF (pH 1.6). The solubility of crystal-line EZ in FaSSGF (pH 1.2) at 25 �C and 37 �C was extremely low:0.2 mg/mL and 0.1mg/mL, respectively (Figure 8(a,b)). Thisextremely low solubility is thought to be due to the un-ionizedstate of the drug in the dissolution medium. In contrast to itsbehavior in FaSSIF, amorphous EZ resulted in no increase in initialsolution concentrations or supersaturation in FaSSGF. In fact, themethod of preparation and the temperature had no effect on this

Figure 7. Dissolution and precipitation (supersaturation) profiles of ternary solid dispersions of ezetimibe (EZ) in fasted-state simulated intestinal fluid: (a) and (b) melt-quenched (MQ1, 2, 3) and spray-dried (SD1, 2, 3) solid dispersions at 25 �C; (c) and (d) MQ and SD solid dispersions at 37 �C. AMP: amorphous EZ; CRY: crystalline EZ.

DRUG DEVELOPMENT AND INDUSTRIAL PHARMACY 7

behavior. This suggests rapid transformation of the amorphousphase into the crystalline phase, possibly as the result of anextremely high driving force for crystallization. Although someincrease in the initial EZ concentrations and supersaturation wereobserved with the MQ and SD solid dispersions in general, the leveland extent of supersaturation were much lower in FaSSGF than inFaSSIF at 25 �C and 37 �C (Figure 8). However, as with FaSSIF,higher polymer/poloxomer content delayed the precipitation kinet-ics, leading to sustained supersaturation for longer periods. Theamorphous form of EZ, which is a poorly soluble and very weaklyacidic drug, can show increased solubility (supersaturation) inFaSSIF and FaSSGF. Although PVP and poloxamer 188 were effect-ive in inhibiting/delaying precipitation from the solid dispersions at25 �C, they were ineffective at 37 �C, which is the physiologicallyrelevant temperature. The method of preparation also affected thesupersaturation and precipitation kinetics. In general, supersatur-ation was maintained for relatively longer in biorelevant media atboth temperatures when MQ solid dispersions were used. It isimportant to consider these aspects in the design and develop-ment of amorphous solid dispersions of poorly soluble drugs.

MDSC, PXRD and FT-IR showed that although we obtainedamorphous solid dispersion formulations, these new systems didnot remain stable when they were dissolved in FaSSIF and FaSSGF,especially for SD formulations at 37 �C.

Conclusions

Ternary solid dispersions of EZ containing PVP and poloxamer weresuccessfully prepared by spray-drying and melt-quenching meth-ods. The powders were thoroughly characterized using solid statetools and were found to be predominantly amorphous in naturewith Tg values in the range of 140–160 �C. FT-IR indicated the

presence of interactions between EZ and the excipients. The resultsindicate that the melting method would be more promising forenhancing the dissolution of EZ from a ternary solid dispersion.Low concentrations of poloxamer 188 are adequate for obtainingthe desired improvements in EZ solubility. The apparent solubilityof EZ solid dispersions was higher in FaSSIF than in FaSSGF andsupersaturation was maintained for an acceptable time, i.e. for thetime required for the absorption process to take place. Further, theprecipitation kinetics were faster at 37 �C than at 25 �C in biorele-vant media, indicating that dissolution studies currently routinelyperformed at 25 �C may be less relevant. In general, MQ solid dis-persions had better supersaturation properties than SD dispersions.This study nicely demonstrated the effects of the formulation, theprocess, the temperature of the dissolution medium, and the typeof biorelevant medium on the supersaturation and precipitationkinetics of ternary solid dispersions of EZ.

Acknowledgements

This research received no specific grant from any funding agencyin the public, commercial or not-for-profit sectors.

Disclosure statement

There were no conflicts of interest.

References

1. Gao P, Shi Y. Characterization of supersaturatable formula-tions for improved absorption of poorly soluble drugs. AAPSJ 2012;14:703–13.

Figure 8. Dissolution and precipitation (supersaturation) profiles of ternary solid dispersions of ezetimibe (EZ) in fasted-state simulated gastric fluid: (a) and (b) melt-quenched (MQ1, 2, 3) and spray-dried (SD1, 2, 3) solid dispersions at 25 �C; (c) and (d) MQ and SD solid dispersions at 37 �C. AMP: amorphous EZ, CRY: crystalline EZ.

8 A. ALHAYALI ET AL.

2. Bevernage J, Brouwers J, Clarysse S, et al. Drug supersatur-ation in simulated and human intestinal fluids representingdifferent nutritional states. J Pharm Sci 2010;99:4525–34.

3. Brouwers J, Brewster ME, Augustijns P. Supersaturating drugdelivery systems: the answer to solubility-limited oral bio-availability? J Pharm Sci 2009;98:2549–72.

4. Dahan A, Beig A, Ioffe-Dahan V, et al. The twofold advantageof the amorphous form as an oral drug delivery practice forlipophilic compounds: Increased apparent solubility anddrug flux through the intestinal membrane. AAPS J2013;15:347–53.

5. Janssens S, Nagels S, De Armas HN, et al. Formulation andcharacterization of ternary solid dispersions made up of itra-conazole and two excipients, TPGS 1000 and PVPVA 64 thatwere selected based on a supersaturation screening study.Eur J Pharm Biopharm 2008;69:158–66.

6. Leuner C, Dressman J. Improving drug solubility for oraldelivery using solid dispersions. Eur J Pharm Biopharm2000;50:47–60.

7. Modi A, Tayade P. Enhancement of dissolution profile bysolid dispersion (kneading) technique. AAPS PharmSciTech2006;7:87–92.

8. Brewster M, Vandecruys R, Verreck G, et al. Supersaturatingdrug delivery systems: effect of hydrophilic cyclodextrinsand other excipients on the formation and stabilization ofsupersaturated drug solutions. Pharmazie 2008;63:217–20.

9. Bevernage J, Hens B, Brouwers J, et al. Supersaturation inhuman gastric fluids. Eur J Pharm Biopharm 2012;81:184–9.

10. Vandecruys R, Peeters J, Verreck G, et al. Use of a screeningmethod to determine excipients which optimize the extentand stability of supersaturated drug solutions and applica-tion of this system to solid formulation design. Int J Pharm2007;342:168–75.

11. Shah TJ, Amin AF, Parikh JR, et al. Process optimization andcharacterization of poloxamer solid dispersions of a poorlywater-soluble drug. AAPS PharmSciTech 2007;8:E18–24.

12. Goddeeris C, Willems T, Van den Mooter G. Formulation offast disintegrating tablets of ternary solid dispersions con-sisting of TPGS 1000 and HPMC 2910 or PVPVA 64 toimprove the dissolution of the anti-HIV drug UC 781. Eur JPharm Sci 2008;34:293–302.

13. Wang X, Michoel A, Van den Mooter G. Solid state character-istics of ternary solid dispersions composed of PVP VA64,Myrj 52 and itraconazole. Int J Pharm 2005;303:54–61.

14. Goddeeris C, Willems T, Houthoofd K, et al. Dissolutionenhancement of the anti-HIV drug UC 781 by formulation ina ternary solid dispersion with TPGS 1000 and EudragitE100. Eur J Pharm Biopharm 2008;70:861–8.

15. Ghebremeskel AN, Vemavarapu C, Lodaya M. Use of surfac-tants as plasticizers in preparing solid dispersions of poorlysoluble API: selection of polymer–surfactant combinationsusing solubility parameters and testing the processability. IntJ Pharm 2007;328:119–29.

16. Shimpi MR, Childs SL, Bostr€om D, et al. New cocrystals ofezetimibe with l-proline and imidazole. CrystEngComm2014;16:8984–93.

17. Gulsun T, Gursoy RN, Oner L. Design and characterization ofnanocrystal formulations containing ezetimibe. Chem PharmBull 2011;59:41–5.

18. Parmar KR, Shah SR, Sheth NR. Studies in dissolutionenhancement of ezetimibe by solid dispersions in combin-ation with a surface adsorbent. Dissolut Technol2011;8:55–61.

19. Taupitz T, Dressman JB, Klein S. New formulation approachesto improve solubility and drug release from fixed dose com-binations: case examples pioglitazone/glimepiride and ezeti-mibe/simvastatin. Eur J Pharm Biopharm 2013;84:208–18.

20. Paradkar A, Ambike AA Jadhav BK, et al. Characterization ofcurcumin–PVP solid dispersion obtained by spray drying. IntJ Pharm 2004;271:281–6.

21. Fousteris E, Tarantili PA, Karavas E, et al. Poly (vinyl pyrroli-done)–poloxamer-188 solid dispersions prepared by hotmelt extrusion. J Therm Anal Calorim 2013;113:1037–47.

22. Soliman S, Abdel Malak N, El Gazayerly O, et al. Preparationof celecoxib solid dispersions for dermal application: in vitrocharacterization and skin irritation test. J Drug Deliv SciTechnol 2011;21:509.

23. Patterson JE, James MB, Forster AH, et al. Preparation ofglass solutions of three poorly water soluble drugs by spraydrying, melt extrusion and ball milling. Int J Pharm2007;336:22–34.

24. Ambike AA, Mahadik K, Paradkar A. Spray-dried amorphoussolid dispersions of simvastatin, a low Tg drug: in vitro andin vivo evaluations. Pharm Res 2005;22:990–8.

25. Zhang Y, Luo R, Chen Y, et al. Application of carrier andplasticizer to improve the dissolution and bioavailability ofpoorly water-soluble baicalein by hot melt extrusion. AAPSPharmSciTech 2014;15:560–8.

26. Newa M, Bhandari KH, Li DX, et al. Preparation, characteriza-tion and in vivo evaluation of ibuprofen binary solid disper-sions with poloxamer 188. Int J Pharm 2007;343:228–37.

27. Paoloni L, Patta A, Mangano F. The hydrogen bond with car-bonyl groups: theoretical study of the correlation betweenthe X–H stretching frequency shift and the C¼O groupproper. J Mol Struct 1975;27:123.

28. Mulye SP, Jamadar SA, Karekar PS, et al. Improvement inphysicochemical properties of ezetimibe using a crystalengineering technique. Powder Technol 2012;222:131–8.

29. Matsumoto T, Zografi G. Physical properties of solid molecu-lar dispersions of indomethacin with poly (vinylpyrrolidone)and poly (vinylpyrrolidone-co-vinyl-acetate) in relation toindomethacin crystallization. Pharm Res 1999;16:1722–8.

30. Khougaz K, Clas S. Crystallization inhibition in solid disper-sions of MK-0591 and poly (vinylpyrrolidone) polymers. JPharm Sci 2000;89:1325–34.

31. Gupta P, Bansal AK. Molecular interactions in celecoxib-PVP-meglumine amorphous system. J Pharm Pharmacol2005;57:303–10.

32. Taylor LS, Zografi G. Spectroscopic characterization of inter-actions between PVP and indomethacin in amorphousmolecular dispersions. Pharm Res 1997;14:1691–8.

33. Doherty C, York P. Evidence for solid- and liquid-state inter-actions in a furosemide-polyvinylpyrrolidone solid dispersion.J Pharm Sci 1987;76:731–7.

34. Carlert S, Pålsson A, Hanisch G, et al. Predicting intestinalprecipitation—a case example for a basic BCS class II drug.Pharm Res 2010;27:2119–30.

35. Amidon GL, Lennern€as H, Shah VP, et al. A theoretical basisfor a biopharmaceutic drug classification: the correlation ofin vitro drug product dissolution and in vivo bioavailability.Pharm Res 1995;12:413–20.

36. Psachoulias D, Vertzoni M, Goumas K, et al. Precipitation inand supersaturation of contents of the upper small intestineafter administration of two weak bases to fasted adults.Pharm Res 2011;28:3145–58.

37. Konno H, Handa T, Alonzo DE, et al. Effect of polymertype on the dissolution profile of amorphous solid

DRUG DEVELOPMENT AND INDUSTRIAL PHARMACY 9

dispersions containing felodipine. Eur J Pharm Biopharm2008;70:493–9.

38. Lindfors L, Forss�en S, Westergren J, et al. Nucleation andcrystal growth in supersaturated solutions of a model drug.J Colloid Interface Sci 2008;325:404–13.

39. Warren DB, Benameur H, Porter CJ, et al. Using polymericprecipitation inhibitors to improve the absorption of poorlywater-soluble drugs: a mechanistic basis for utility. J DrugTarget 2010;18:704–31.

40. Usui F, Maeda K, Kusai A, et al. Inhibitory effects of water-soluble polymers on precipitation of RS-8359. Int J Pharm1997;154:59–66.

41. Bevernage J, Brouwers J, Brewster ME, et al. Evaluation ofgastrointestinal drug supersaturation and precipitation: strat-egies and issues. Int J Pharm 2013;453:25–35.

42. Greco K, Bogner R. Crystallization of amorphous indometh-acin during dissolution: effect of processing and annealing.Mol Pharm 2010;7:1406–18.

10 A. ALHAYALI ET AL.

SUPPLEMENTARY DATA

Figure S1. Supplementary Figure, Thermogravimetric analysis curves of ezetimibe (EZ), poloxamer 188 and polyvinylpyrrolidone (PVP) K-30 showing the degradation temperatures of the main components of the solid dispersion formulations.

Paper II

Contents lists available at ScienceDirect

European Journal of Pharmaceutical Sciences

journal homepage: www.elsevier.com/locate/ejps

Investigation of supersaturation and in vitro permeation of the poorly watersoluble drug ezetimibe

Amani Alhayalia, Mohammed Ali Selob,c, Carsten Ehrhardtb, Sitaram Velagaa,⁎

a Pharmaceutical Research Group, Department of Health Sciences, Luleå University of Technology, Luleå, Swedenb School of Pharmacy and Pharmaceutical Sciences and Trinity Biomedical Sciences Institute, Trinity College Dublin, Dublin, Irelandc Faculty of Pharmacy, University of Kufa, Al-Najaf, Iraq

A R T I C L E I N F O

Keywords:Poorly soluble drugsBiorelevantSupersaturationPrecipitationCaco-2 cellsAbsorption

A B S T R A C T

The interplay between supersaturation, precipitation and permeation characteristics of the poorly water-solubledrug ezetimibe (EZ) was investigated. Supersaturation and precipitation characteristics of EZ in the presence ofCaco-2 cells were compared to those in a cell-free environment. The effect of the water-soluble polymer poly-vinyl pyrrolidone (PVP-K30) on the supersaturation, precipitation and transport of EZ was also investigated andthe amount of drug taken up by Caco-2 cells was quantified.

A one-compartment setup without Caco-2 cells (i.e. in the wells of cell-culture plates) was used to mimic anon-sink in vitro dissolution chamber. The two-compartment Caco-2 cell monolayer setup (with apical andbasolateral compartments) was used to investigate how the absorption of EZ affects supersaturation. EZ invarying degrees of supersaturation (DS; 10, 20, 30 and 40) was introduced into the one-compartment setup orthe apical chamber of the two-compartment setup. Samples were collected at specific times to determine su-persaturation, precipitation and permeation. At the end of the study, Caco-2 cells were lysed and the in-tracellular amount of EZ was quantified.

In the one-compartment setup, a high DS was associated with rapid precipitation. Supersaturation wasmaintained for longer time periods and precipitation was lower in the presence of Caco-2 cells. There were nosignificant differences in the absorption rate of the drug, even at high concentrations on the apical side.Permeability coefficients for all supersaturated solutions (i.e. DS 10–40) were significantly (p < 0.05) differentfrom those when EZ was present in crystalline form. Both concentrations of PVP-K30 (i.e. 0.05% and 0.1% w/v)improved solubility and supersaturation of EZ when added to the apical side, however, the increase in absorptionat the higher concentration was not proportional. The amount of intracellular EZ increased with increasing DS inthe apical side, until the saturation limit was reached in the cells (i.e. at DS 30 and higher).

This study demonstrated that precipitation of EZ could be overestimated when supersaturation was in-vestigated without the implementation of an absorption compartment in vitro, both in the absence and in thepresence of polymer.

1. Introduction

Development of new pharmaceutical products for oral administra-tion has been challenged by the increasing number of poorly water-soluble molecules in the pipeline. Most of these drug moleculespermeate well through the intestinal wall despite their poor solubilityin gastrointestinal fluids [i.e. Biopharmaceutical Classification System(BCS) class II compounds] (Amidon et al., 1995; Frank et al., 2014). Thesolubility and permeation of drugs are key characteristics that de-termine their oral bioavailability; the poor solubility of class II com-pounds restricts their absorption and hence their systemic

bioavailability. Drugs need to dissolve and be in solution in the in-testinal lumen to be available for absorption through the gut wall.Various conventional and advanced approaches have been used byformulation scientists to improve the solubility, and thus the bioavail-ability, of these drug molecules (Bikiaris, 2011; Singh et al., 2011).

Supersaturating drug delivery systems (SDDSs) such as solid dis-persions are one approach to improve solubility of orally administereddrugs (Alonzo et al., 2011). SDDSs increase the aqueous solubility ofpoorly soluble drugs by creating a supersaturation state in the gastro-intestinal lumen, thus improving bioavailability. A supersaturationstate can be created either intentionally, using drug formulation

https://doi.org/10.1016/j.ejps.2018.01.047Received 20 November 2017; Received in revised form 30 January 2018; Accepted 31 January 2018

⁎ Corresponding author at: Department of Health Sciences, Division of Medical Sciences, Lulea University of Technology, SE-971 87 Luleå, Sweden.E-mail address: sitaram.velaga@ltu.se (S. Velaga).

strategies such as SDDSs, or naturally after the passage of weakly basiccompounds from the stomach into the small intestine (Brouwers et al.,2009). The drug molecules in a supersaturated state tend to precipitaterapidly, as this is a thermodynamically unstable state. Thus, the im-provements to solubility and absorption achieved by SDDSs will dependon the extent of the supersaturation and the duration of maintaining thesupersaturation state in the gastrointestinal lumen. The challenge is increating concentrations of the drug that are many times higher than itsthermodynamic solubility concentration, and maintaining these highconcentrations for an adequate time period for the absorption processto take place (Alonzo et al., 2011; Hou and Siegel, 2006).

The supersaturated state has predominantly been investigated invitro, as studying supersaturation in vivo is difficult (Gao et al., 2004;Bevernage et al., 2013; Alhayali et al., 2016). Most of these studies havebeen carried out in one-compartment systems, which neglect the effectof intestinal absorption on supersaturation behaviour. Absorption canaffect supersaturation/precipitation of the drug in the intestinal lumenas it can relieve the unstable supersaturated state and prioritise drugpermeation over precipitation. Therefore, the prediction power of thedissolution and precipitation kinetics increases when an acceptor ab-sorption compartment is added to the dissolution evaluation setup (Shiet al., 2010; Bevernage et al., 2012).

Caco-2 cell monolayers have become the gold standard of intestinaldrug disposition studies (Van Breemen and Li, 2005). A two-compart-ment setup [with apical (AP) and basolateral (BL) sides] was first im-plemented as a successful predictive tool for measuring the dissolutionand absorption of drugs simultaneously by Kataoka et al. (2012, 2003).In these studies, Caco-2 cells were used as a middle layer between thedissolution (donor) and absorption (acceptor) chambers. The systemwas useful for predicting the oral absorption of poorly water-solubledrugs after administration as solid dosage forms. Later, Bevernage et al.(2012) employed a two-compartment set-up with Caco-2 cell mono-layer to investigate the supersaturation behaviour of Loviride to un-derstand the interplay among supersaturation-precipitation-permeation(Bevernage et al., 2012). Frank et al. (2012) has studied the relationshipbetween the true supersaturation and apparent solubility of amorphoussolid dispersions of ABT-102.

Ezetimibe (EZ) (molecular weight 409.4 gmole−1) is a poorlywater-soluble drug and belongs to BCS (class II). EZ is weakly acidcompound with pKa of 9.75 and log p value of 1.6 (Srivalli and Mishra,2016). In our previous study, the supersaturation and precipitationbehaviour of solid dispersions of EZ was investigated in one-compart-ment using biorelevant media (Alhayali et al., 2016). Absorption pro-cess across the intestinal epithelial barrier could influence considerablythe supersaturation and precipitation behaviour. In this study, the effectof absorption (across Caco-2 cell monolayers) on the supersaturationand precipitation of EZ was investigated in two compartments andcompared to the supersaturation behaviour in a cell-free, one-com-partment setup. Furthermore, the effect of the hydrophilic polymerPVP-K30 (polyvinylpyrrolidone) was also investigated. PVP-K30 a hy-drophilic polymer is widely and commonly used as supersaturationinducing-agent and precipitation inhibitor for the formulation of poorlysoluble drugs (Patel et al., 2008; Xu et al., 2008; Jahangiri et al., 2016).

2. Materials and methods

2.1. Materials

EZ, Dulbecco's modified Eagle's Medium (DMEM), HEPES, Triton X-100, DMSO, NaOH pellets, KH2-phosphate and NaCl were purchasedfrom Sigma-Aldrich (Stockholm, Sweden); simulated intestinal fluid(SIF) powder for the preparation of fasted-state SIF (FaSSIF) was pur-chased from biorelevant.com (Surrey, United Kingdom) and Hanksbalanced salt solution (HBSS) was obtained from Gibco (Paisley, UnitedKingdom). Tocopheryl polyethylene glycol succinate (TPGS) and PVP-K30 were received as free samples from BASF chemicals (Ludwigshafen,

Germany).All solvents used for HPLC analysis were of chromatography grade.

Pure Millipore water was used.

2.2. Cell culture

Caco-2 cells were routinely cultured in 75 cm2 tissue culture flasks(Greiner Bio One, Frickenhausen, Germany) at 37 °C in 5% CO2 atmo-sphere in high glucose DMEM supplemented with 20% fetal bovineserum (FBS), 1 mM sodium pyruvate, 100 U/mL penicillin and 100 μg/mL streptomycin. For transport studies, the cells were seeded at adensity of 75,000 cells/cm2 on rat-tail collagen type I-coated Transwellclear inserts (12mm in diameter, pore size 0.4 μm; Corning, VWR,Dublin, Ireland). The cells were allowed to grow for at least 16 days andthe medium was exchanged every other day (Frank et al., 2014). Thetransepithelial electrical resistance (TEER) of cell monolayers wasmeasured using a Millicell ERS-2 epithelial volt-ohm meter equippedwith STX-2 electrodes (Millipore, Carrigtwohill, Ireland) and correctedfor the background value contributed by the cell culture inserts andmedium.

2.3. Membrane integrity measurements

All sample solutions under study were investigated for toxic effectson the integrity of the cells before starting the transport experiments.After aspiration of the culture medium, insert-grown Caco-2 cellmonolayers were washed with pre-warmed HBSS (37 °C), then HBSSwas added to the AP and BL sides and cells were incubated at 37 °C for1 h. Subsequently, TEER values were measured and used as controlvalues for each sample solution under study. Afterward, the HBSS in theAP compartment of all inserts was replaced with FaSSIF containingdifferent concentrations of EZ or EZ with polymer (the experimentalvariables under study) and cells were incubated for an additional 2 h. Atthe end of the incubation period, sample solutions were removed andHBSS was added to both AP and BL sides and TEER values were mea-sured. TEER % was calculated as (TEER at 2 h/TEER at time 0)× 100.These values were used to select samples for the transport experiments.

In addition to these measurements, to exclude any effects of thesampling process on the cells' integrity and to ensure the tightness ofthe inter-cellular junctions (Yee, 1997), the TEER values for all cellmonolayers were measured before starting and at the end of each ex-periment. Before starting the experiments, DMEM was removed fromthe AP and BL compartments, the Caco-2 cells were incubated withHBSS (37 °C) for 1 to 1.5 h and the TEER values were measured. At theend of the experiment (after 2 h), the experimental medium in the APand BL compartments was replaced with HBSS and the TEER valueswere measured.

2.4. Supersaturation-precipitation-permeation interplay (without polymer)

2.4.1. One-compartment setupA one-compartment experimental setup (without Caco-2 cells) was

used to investigate the supersaturation and precipitation behaviour ofEZ in the absence of a Caco-2 cell monolayer. The supersaturation ex-periments were conducted in 12-wells plates (22.1 mm diameter).FaSSIF (0.5 mL) was added to each well and supersaturation was in-duced using the solvent shift method starting with a higher con-centration of DMSO stock solution. The final concentration of DMSO inFaSSIF was<1%. Different degrees of supersaturation (DS) (DS 10, DS20, DS 30 and DS 40) were induced based on the equilibrium con-centration of EZ in FaSSIF. For example, the concentration of the so-lution at DS10 is 10 times the equilibrium solubility of EZ. The otherexperimental conditions were as follows; 60 rpm shaking speed and37 °C temperature. Samples were withdrawn at fixed time points andcentrifuged immediately, and the obtained supernatants were dilutedand analysed using HPLC.

A. Alhayali et al.

2.4.2. Two-compartment setup2.4.2.1. Apical compartment (donor compartment). FaSSIF was used asthe supersaturation medium in the AP compartment. It was preparedfrom the original SIF powder using potassium di‑hydrogen phosphatebuffer with the pH adjusted to 6.5 with NaOH. The solvent shift methodwas used and different DS (DS 10, DS 20, DS 30 and DS 40) wereinduced as in Section 2.4.1. The other experimental details can be foundin the Section 2.4.3.

2.4.2.2. Basolateral compartment (acceptor compartment). HBSS was thebasis of the transport medium in the BL compartment of the cellmonolayer. The HBSS was buffered with HEPES 10mM to a pH of 7.4and supplemented with 25mM glucose. TPGS was added to thetransport medium in a concentration of 0.2% to provide sinkconditions (Hou and Siegel 2006). This concentration of TPGS waschosen following solubility analysis of the drug in the medium and aftertesting for compatibility with Caco2 cells (data not shown). The otherexperimental details can be found in the Section 2.4.3.

2.4.3. Vectorial transport experimentsExperiments for the investigation of transport were conducted in the

presence of Caco-2 cell monolayers using Transwell inserts. FaSSIF(0.5 mL) was prepared and added to the AP side and different DSs wereinduced as outlined before. Samples from AP (40 μL) were taken at 0, 5,15, 30, 45, 60, 90 and 120min and centrifuged immediately at 17,000 gfor 5–10min. The supernatants were diluted with the mobile phase andanalysed using HPLC to find the concentration of EZ. The amount of EZin the apical compartment (0.5 mL) was quantified.

Two-hundred microliter samples were withdrawn from the BL sideat the same time points and analysed without dilution. Fresh buffer(200 μL) was added to the BL side at each time point to ensure themaintenance of sink conditions.

All experiments were conducted in triplicate. Each plate set oftranswell membrane inserts was kept in an agitating incubator with anorbital shaker (speed= 60 rpm and medium temperature= 37 °C) andwas only taken out for sampling.

2.5. Supersaturation-precipitation-permeation interplay (with polymer)

The effect of the polymer on supersaturation and permeation wasstudied in one-compartment and two compartment setups (as shown inSections 2.4.1 and 2.4.2.) with the DS 40 sample. PVP K-30 was pre-dissolved in FaSSIF in two concentrations: 0.05 and 0.1% w/v in the APside of the two compartment setup and in the one-compartment setup.The DS 40 supersaturated solution of EZ was prepared as outlined inSection 2.4 and was introduced into the FaSSIF solutions.

2.6. Quantification of intracellular ezetimibe in Caco-2 cells

At the end of transport studies, the Caco-2 cell monolayers werewashed twice with HBSS and lysed in 200 μL of FaSSIF supplementedwith 1% Triton X-100. After 2–3min, cell monolayers were removedgently from the surface of the insert membranes, diluted with acet-onitrile and analysed using HPLC.

2.7. Quantitative analysis of ezetimibe

An isocratic high performance liquid chromatography (HPLC)method was used for the determination of EZ concentrations in thecollected samples. An Agilent HPLC unit (1290 series) equipped with aquaternary pump and UV detector (Agilent Technologies, Waldbronn,Germany) and a C-18 Eclipse column with a 5 μm (4.6mm×100mm)were used. A 40:60 (v/v) mobile phase was prepared from 0.05% w/vsodium 1-heptane sulfonic acid and acetonitrile. A flow rate of 0.5 mL/min and injection volume of 10 μL were used. The measurements weremade at a wavelength of 230 nm.

2.8. Statistical analysis

One-way ANOVA and a t-test were used to investigate statisticallysignificant (p < 0.05) differences in the data obtained. All experimentswere carried out in triplicate (n= 3).

3. Results and discussion

3.1. Integrity of Caco-2 cell monolayers

The TEER values were measured to determine the integrity of thecell monolayer on exposure to the different drug solutions under study(Fig. 1). The results were given as percentage of the initial TEER value.All TEER % values were either stable or increased compared to thecontrol TEER value (i.e.≥ 100%). Samples containing PVP-K30 re-sulted in a decreased values (approximately 80%) but always remainedabove the threshold of 300Ω×cm2, which generally is considered tobe a hallmark of Caco-2 cell monolayer integrity (Van Breemen and Li,2005; Frank et al., 2012; Saha and Kou, 2000). The TEER values beforeand after each experiment were high (at least 1700Ω×cm2) indicatingthat the cell integrity was kept intact until the end of the experiment.

3.2. Supersaturation-precipitation-permeation interplay (without polymer)

Supersaturation experiments were initially conducted in wells ofcell culture plates (one-compartment setup) to gain insight into thesupersaturation/precipitation behaviour of EZ in the absence of Caco-2cells (i.e. in a cell-free environment) and to understand whether su-persaturation/precipitation experiments conducted in a one-compart-ment model or in a test tube can predict the behaviour of a drugcompound in a two-compartment model.

As illustrated in Fig. 2a, the EZ concentration was higher with theDS 10 sample than with crystalline EZ and supersaturation was main-tained until the end of the experiment. EZ precipitated with the DS 20and DS 30 samples after 15 and 5min, respectively. Instantaneousprecipitation occurred with the DS 40 sample in FaSSIF (Fig. 2a). In thisstudy, spontaneous nucleation and precipitation of EZ were observed inFaSSIF in all supersaturated solutions, suggesting a low kinetic barrierto crystallisation of the drug. This is typical behaviour because thesupersaturated state is thermodynamically unstable and faster pre-cipitation occurs as supersaturation increases, according to the theoryof crystal growth (Lindfors et al., 2008). In contrast, Bevernage et al.(2012) observed no spontaneous precipitation and nuclei were requiredto promote precipitation (Bevernage et al., 2012).

Fig. 1. Changes in normalised transepithelial electrical resistance (TEER) following apicalincubation of Caco-2 cell monolayers with various samples. Incubation with HBSS wasused as control. Data represent means ± SD (n=4). DS=degree of supersaturation ofezetimibe.

A. Alhayali et al.

Fig. 2b shows the total amounts of EZ in FaSSIF on the AP sidewithout addition of polymer. As depicted, the amount of drug dissolvedwas revealed to increase as the DS increased. All DS levels were

associated with more stable solutions with lower precipitation ten-dencies. The reason for this could be that the permeation of the drugwas an alternative to precipitation for relieving the thermodynamicallyunstable state that is the predominant feature during supersaturation,and as the concentration (or the DS) increased, the possibility of per-meation reducing precipitation also increased, as reported byBevernage et al. (2012).

The permeation of EZ across the Caco-2 cell barrier at differentconcentrations was presented as a flux (J) instead of as the apparentpermeability (Papp), because reliable Papp measurements require thatthe donor concentrations are known and determination of the drug fluxacross the membranes is independent of the donor concentration. Theexact initial concentration of the dissolved drug was not known becauseof the fast precipitation of poorly water-soluble drugs when super-saturation was induced. This has also been documented for other drugsin previous studies (Kanzer et al., 2010; Hubatsch et al., 2007).

The effect of inducing different degrees of supersaturation on thepermeation of EZ across the Caco-2 cell monolayer is shown in Fig. 2c.The cumulative amounts of EZ in the transport medium (HBSS) in theBL compartment is plotted as a function of time. As delineated inFig. 2c, the transport of drug to the BL side was delayed for about45min with DS 10 and 30min with DS 20, 30 and 40. This lag-timecould be due to the time needed to saturate the monolayer barrier be-fore reaching steady-state flux, as reported by Kanzer et al. (2010). Thehighest amount of EZ transported to the BL compartment after 2 h was1.28 ± 1.1E-04 μg from the DS 40 sample; 40.83 ± 2.27 μg was insolution in the AP compartment after 2 h, as shown in Fig. 2b. Theresulting flux was 0.012 ± 9.9E-04 μg/min× cm2 (Fig. 3). Although

Fig. 2. Amount of ezetimibe (EZ) in FaSSIF in the one-compartment setup (a). Amount of EZ in FaSSIF on the apical side of Caco-2 cell monolayers as a function of time, after induction ofsupersaturation (b). Cumulative amount of EZ recovered from the basolateral acceptor medium (i.e. HBSS), after vectorial transport across Caco-2 cell monolayers upon induction ofsupersaturation in the apical compartment (c). Data represent means ± SD (n=3). Error bars in (c) are too small to show.

Fig. 3. Calculated flux of various ezetimibe (EZ) formulations across Transwell-grownCaco-2 cell monolayers. Data represent means ± SD (n=3). Symbol (*) represent thesignificant difference in the flux of the different polymer concentrations (p < 0.05), (ns)indicates data is not significant with respect to the flux of all DSs (p > 0.05).

A. Alhayali et al.

the DS 10 sample provided the highest amount of EZ dissolved after 2 hin the one-compartment setup (Fig. 2a), the highest amount transportedafter 2 h in the two-compartment setup was from the DS 40 sample. TheDS 40 sample also maintained the highest amount of dissolved drug inthe supersaturated state on the AP side of the two-compartment setup(Fig. 2a and b). Differences between the DS samples in the amounts ofEZ transported to the BL side were less pronounced than differences inthe precipitation behaviour in the one-compartment setup. However,precipitation behaviour in the one-compartment setup was not in ac-cordance with that on the AP side in the two-compartment setup. Thedifferent outcomes for the one-compartment and two-compartmentsetups demonstrate that reliable prediction of supersaturation/pre-cipitation behaviour cannot be achieved without including tools topredict absorption in the in vitro setup, as reported by Bevernage et al.(2012). As shown in Fig. 3, there were no significant differences in fluxvalues between samples with DS 10, 20, 30 or 40 for drug transport tothe BL side (p > 0.05). However, the EZ fluxes with DS 10, 20, 30 and40 samples differed significantly from the flux with the crystalline EZsample (p < 0.05). As reported in a previous study (Frank et al., 2012),the amount of drug transported to the BL side depends on the amount ofdrug in the supersaturated solution on the AP side. The specific beha-viour of EZ as it was taken up and metabolised by enterocytes couldexplain the lack of effect of the varying DSs on EZ flux on the apical sideof the two-compartment setup, as reported by Oswald et al. (2007).Generally, poorly water soluble drugs that permeate the intestinal wallwell should show increased permeation when the concentration of thedissolved drug is increased on the AP side, but some recent reports haveshown that solubilised colloidal drugs may permeate poorly (Franket al., 2012; Fischer et al., 2011).

3.3. Supersaturation-precipitation-permeation interplay (with polymer)

PVP-K30 was selected to assess the effect of polymers on the in-hibition of crystallisation (i.e. on sustaining supersaturation) andtransport. PVP K-30 is well established as a precipitation inhibitor(Alhayali et al., 2016; Patel and Anderson, 2015).

In the one-compartment setup with the DS 40 sample (Fig. 4a), aPVP-K30 concentration of 0.05% w/v resulted in a higher amount ofdrug in solution but precipitation occurred gradually over 90min, atwhich point all the drug had been precipitated. A PVP-K30 concentra-tion of 0.1% w/v resulted in rapid precipitation over the first 30minbut did not perform better than 0.05% w/v on keeping supersaturationstate for longer time. Precipitation occurred later with both polymerconcentrations (0.05% and 0.1% w/v) than in the absence of polymer.In the two-compartment setup, a concentration of 0.05% w/v main-tained the supersaturation for the 2 h (Fig. 4b) while it was difficult toconclude the effect of 0.1% w/v due to system variability.

In general, less extensive precipitation was observed in the two-compartment setup, and the amount of EZ dissolved on the AP side wasgreater in the presence of PVP-K30 (Fig.4b). The polymer PVP-K30(0.05% w/v) kept the drug in the supersaturated state for 2 h. In theabsence of polymer, the drug started to precipitate after 15min. On theother hand, PVP-K30 (0.1% w/v) in the medium on the AP side in-creased the amount of dissolved drug up to 20 μg after 5min, whichwas more than with a concentration of 0.05% w/v and in the absence ofpolymer (Fig. 4b). The effect of PVP-K30 on the supersaturation of EZ inFaSSIF has been demonstrated previously (Alhayali et al., 2016). Thereason for less extensive precipitation of EZ in the presence of Caco-2cells may be that permeation offered an alternative to precipitation. Thesame conclusion applies in the absence of polymer. Thus the in vitroprediction of supersaturation and precipitation is more accurate if it iscombined with absorption tools because, in some cases, such as in thisstudy, the precipitation can be overestimated by the one-compartmentexperiments compared to the two-compartment experiments.

In this study, we also investigated how supersaturation affects thetransport of EZ in vitro in a two-compartment setup using Caco2 cells in

the presence of PVP-K30. Fig. 4c shows the cumulative amounts of EZon the BL side that were transported over 2 h. Addition of 0.05% w/vPVP-K30 to the AP side resulted in the highest amount of EZ trans-ported to the BL side. Furthermore the flux of EZ was significantlyimproved (p < 0.05) with this polymer concentration 0.064 ± 1.5E-02 μg/min× cm as compared to 0.1% w/v polymer and no polymer(Fig. 3). PVP-K30 (0.1% w/v) improved the supersaturation solubilityof EZ on the apical side in FaSSIF without improving the flux; the totaltransported amount was 2.97 ± 7.5E-04 μg.

In summary, PVP-K30 (0.05% w/v) on the AP side in FaSSIF keptthe drug in the supersaturated state for 2 h, improved the flux, andachieved the highest cumulated EZ concentration on the BL side after2 h (5.54 μg ± 2.6E-03), as shown in Fig. 4b&c. The increase in drugconcentration on the donor AP side as a result of solubilizing the ex-cipients may not cause a proportional improvement in the flux in allcases, as reported by Saha and Kou, 2000. The intrinsic physicochem-ical properties of the drug and drug-excipient interaction can be ex-pected to affect the outcome. In this study, PVP-K30 improved the fluxacross the Caco-2 cell barrier (Fig. 3) without affecting the amount ofEZ taken up by the monolayer, as shown in Fig. 5. However, some re-cent studies have shown the opposite effect, reporting that addition ofsome excipient to the AP side has a negative effect on transport. Thiswas attributed to the fact that increased apparent solubility is asso-ciated with a micellar solubilisation effect on the excipients which canresult in overestimation of the true supersaturation, thus not causingimprovement in the permeation rate (Frank et al., 2012).

3.4. Quantification of intracellular ezetimibe

EZ is known to accumulate in the intestinal mucosa (Oswald et al.,2007). To investigate how much of the EZ was taken up by the Caco-2cells, the cells were lysed and analysed for intracellular EZ using HPLC.Fig. 5 shows the amounts of EZ in each sample solution studied. WithDS 10, 7 ± 3.02 μg of EZ was found in the cell monolayer; this wasapprox. 4.3-fold the amount for the crystalline form. The amounts withDS 20, 30 and 40 were 13 ± 0.99, 32.7 ± 5.80 and 26.8 ± 6.70 μg,approximately 8-, 20- and 16-fold, respectively. These results indicatethat the uptake of EZ by the cells increases proportionally with thedegree of supersaturation on the AP side, reaching a plateau when theconcentration of the drug is 30 times the concentration of the crystal-line form at equilibrium, or higher (DS 30 and 40), on the AP donorside. The uptake of the drug by Caco-2 cells from the DS 40 sample inthe presence of PVP-K30 (at 0.05 and 0.1% w/v) was approximately16.7- and 20-fold the amount quantified for the crystalline form, re-spectively. These amounts were the same as in the absence of thepolymer, indicating that the amount taken up by the cells was not af-fected by the presence of polymer. Further, the uptake of EZ by Caco2cells reached saturation for samples with DS 30 and higher, both withand without polymer. In previous work by Saha and Kou (2000), it wasreported that the addition of solubilizing excipients to the AP donorside of Caco-2 cells may not only improve the solubility of a poorlysoluble drug, but also affect the monolayer drug concentration and thecumulated drug amount, which was in contrast to our findings for EZ.To the best of our knowledge, this is the first study which has quantifiedthe amount of EZ in Caco-2 cells during supersaturation in FaSSIF.

4. Conclusions

This study was an attempt to investigate the supersaturation-pre-cipitation-permeation interplay of EZ in a Caco-2 cell environment. Thesupersaturated solutions of EZ showed higher precipitation earlier atearly time points in one compartment chamber while less precipitationtendency had been seen in the presence of Caco-2 cells. The transport ofEZ for all degrees of supersaturation was low. Thus, the supersaturationstudies that are usually conducted in one compartment chamber or intest tube could underestimate supersaturation, precipitation and

A. Alhayali et al.

transport of the drug. The polymer PVP-K30 improved the super-saturation of EZ at concentrations at 0.05% w/v and resulted in sig-nificant improvement in the transport/flux (J). EZ was taken up by theCaco-2 cells, with the amount trapped increasing with increasingamounts of the drug on the AP side. Hence, a more in vivo-like tool is

mandatory for accurate prediction of bioavailability.

Acknowledgments

A.A. and M.A.S. are recipients of PhD bursaries from the IraqiMinistry of Higher Education and Scientific Research (MOHESR); theauthors would like to acknowledge their contribution and funding.

References

Alhayali, A., Tavellin, S., Velaga, S., 2016. Dissolution and precipitation behavior ofternary solid dispersions of ezetimibe in biorelevant media. Drug Dev. Ind. Pharm.1–10.

Alonzo, D.E., Gao, Y., Zhou, D., Mo, H., Zhang, G.G., Taylor, L.S., 2011. Dissolution andprecipitation behavior of amorphous solid dispersions. J. Pharm. Sci. 100,3316–3331.

Amidon, G.L., Lennernäs, H., Shah, V.P., Crison, J.R., 1995. A theoretical basis for abiopharmaceutic drug classification: the correlation of in vitro drug product dis-solution and in vivo bioavailability. Pharm. Res. 12, 413–420.

Bevernage, J., Brouwers, J., Annaert, P., Augustijns, P., 2012. Drug pre-cipitation–permeation interplay: supersaturation in an absorptive environment. Eur.J. Pharm. Biopharm. 82, 424–428.

Bevernage, J., Brouwers, J., Brewster, M.E., Augustijns, P., 2013. Evaluation of gastro-intestinal drug supersaturation and precipitation: strategies and issues. Int. J. Pharm.453, 25–35.

Bikiaris, D.N., 2011. Solid dispersions, part I: recent evolutions and future opportunitiesin manufacturing methods for dissolution rate enhancement of poorly water-solubledrugs. Expert Opin. Drug Deliv. 8, 1501–1519.

Brouwers, J., Brewster, M.E., Augustijns, P., 2009. Supersaturating drug delivery systems:the answer to solubility-limited oral bioavailability? J. Pharm. Sci. 98, 2549–2572.

Fischer, S.M., Brandl, M., Fricker, G., 2011. Effect of the non-ionic surfactant Poloxamer188 on passive permeability of poorly soluble drugs across Caco-2 cell monolayers.Eur. J. Pharm. Biopharm. 79, 416–422.

Fig. 4. Amount of ezetimibe (EZ) with PVP-K30 in FaSSIF in the one-compartment setup (a). Amount of EZ with PVP-K30 in FaSSIF on the apical side of Caco-2 cell monolayers as afunction of time, after induction of supersaturation to DS 40 (b). Cumulative amount of EZ recovered from the basolateral acceptor medium (i.e. HBSS) after vectorial transport acrossCaco-2 cell monolayers upon induction of supersaturation in the presence of polymer in the apical compartment (c). Data represent means ± SD (n=3). Error bars in (c) are too small toshow. DS=degree of saturation.

Fig. 5. Amount of intracellular ezetimibe (EZ) after transmonolayer transport acrossCaco-2 cells. Data represent means ± SD (n=3).

A. Alhayali et al.

Frank, K.J., Rosenblatt, K.M., Westedt, U., Hölig, P., Rosenberg, J., Mägerlein, M., Fricker,G., Brandl, M., 2012. Amorphous solid dispersion enhances permeation of poorlysoluble ABT-102: true supersaturation vs. apparent solubility enhancement. Int. J.Pharm. 437, 288–293.

Frank, K.J., Westedt, U., Rosenblatt, K.M., Hölig, P., Rosenberg, J., Mägerlein, M., Fricker,G., Brandl, M., 2014. What is the mechanism behind increased permeation rate of apoorly soluble drug from aqueous dispersions of an amorphous solid dispersion? J.Pharm. Sci. 103, 1779–1786.

Gao, P., Guyton, M.E., Huang, T., Bauer, J.M., Stefanski, K.J., Lu, Q., 2004. Enhanced oralbioavailability of a poorly water soluble drug PNU-91325 by supersaturatable for-mulations. Drug Dev. Ind. Pharm. 30, 221–229.

Hou, H., Siegel, R.A., 2006. Enhanced permeation of diazepam through artificial mem-branes from supersaturated solutions. J. Pharm. Sci. 95, 896–905.

Hubatsch, I., Ragnarsson, E.G., Artursson, P., 2007. Determination of drug permeabilityand prediction of drug absorption in Caco-2 monolayers. Nat. Protoc. 2, 2111–2119.

Jahangiri, A., Barzegar-Jalali, M., Garjani, A., Javadzadeh, Y., Hamishehkar, H.,Asadpour-Zeynali, K., Adibkia, K., 2016. Evaluation of physicochemical propertiesand in vivo efficiency of atorvastatin calcium/ezetimibe solid dispersions. Eur. J.Pharm. Sci. 82, 21–30.

Kanzer, J., Tho, I., Flaten, G.E., Mägerlein, M., Hölig, P., Fricker, G., Brandl, M., 2010. In-vitro permeability screening of melt extrudate formulations containing poorly water-soluble drug compounds using the phospholipid vesicle-based barrier. J. Pharm.Pharmacol. 62, 1591–1598.

Kataoka, M., Masaoka, Y., Yamazaki, Y., Sakane, T., Sezaki, H., Yamashita, S., 2003. Invitro system to evaluate oral absorption of poorly water-soluble drugs: simultaneousanalysis on dissolution and permeation of drugs. Pharm. Res. 20, 1674–1680.

Kataoka, M., Sugano, K., da Costa Mathews, C., Wong, J.W., Jones, K.L., Masaoka, Y.,Sakuma, S., Yamashita, S., 2012. Application of dissolution/permeation system forevaluation of formulation effect on oral absorption of poorly water-soluble drugs in

drug development. Pharm. Res. 29, 1485–1494.Lindfors, L., Forssén, S., Westergren, J., Olsson, U., 2008. Nucleation and crystal growth

in supersaturated solutions of a model drug. J. Colloid Interface Sci. 325, 404–413.Oswald, S., Koll, C., Siegmund, W., 2007. Disposition of the cholesterol absorption in-

hibitor ezetimibe in mdr1a/b (−/−) mice. J. Pharm. Sci. 96, 3478–3484.Patel, D.D., Anderson, B.D., 2015. Adsorption of Polyvinylpyrrolidone and its impact on

maintenance of aqueous Supersaturation of indomethacin via crystal growth inhibi-tion. J. Pharm. Sci. 104, 2923–2933.

Patel, R.P., Patel, D.J., Bhimani, D.B., Patel, J.K., 2008. Physicochemical characterizationand dissolution study of solid dispersions of furosemide with polyethylene glycol6000 and polyvinylpyrrolidone K30. Dissolut. Technol. 15, 17–25.

Saha, P., Kou, J.H., 2000. Effect of solubilizing excipients on permeation of poorly water-soluble compounds across Caco-2 cell monolayers. Eur. J. Pharm. Biopharm. 50,403–411.

Shi, Y., Gao, P., Gong, Y., Ping, H., 2010. Application of a biphasic test for character-ization of in vitro drug release of immediate release formulations of celecoxib and itsrelevance to in vivo absorption. Mol. Pharm. 7, 1458–1465.

Singh, A., Worku, Z.A., Van den Mooter, G., 2011. Oral formulation strategies to improvesolubility of poorly water-soluble drugs. Expert Opin. Drug Deliv. 8, 1361–1378.

Srivalli, K.M.R., Mishra, B., 2016. Improved aqueous solubility and anti-hypercholesterolemic activity of ezetimibe on formulating with hydroxypropyl-β-cyclodextrin and hydrophilic auxiliary substances. AAPS PharmSciTech 17, 272–283.

Van Breemen, R.B., Li, Y., 2005. Caco-2 cell permeability assays to measure drug ab-sorption. Expert Opin. Drug Deliv. 1, 175–185.

Xu, D.H., Wang, S., Mei, X.T., Luo, X.J., Xu, S.B., 2008. Studies on solubility enhancementof curcumin by Polyvinylpyrrolidione K30. Zhong Yao Cai 31, 438–442.

Yee, S., 1997. In vitro permeability across Caco-2 cells (colonic) can predict in vivo (smallintestinal) absorption in man—fact or myth. Pharm. Res. 14, 763–766.

A. Alhayali et al.

Paper III

1

Cocrystal of tadalafil: Physicochemical characterization,

pH-solubility and supersaturation studies Manishkumar R. Shimpi1,2, Amani Alhayali1, Katie L. Cavanagh3, Naír Rodríguez-

Hornedo3, Sitaram P. Velaga1, *

1Pharmaceutical and Biomaterials group, Department of Health Sciences 2Chemistry of Interfaces, Luleå University of Technology, SE-97187, Luleå, Sweden.

3Department of Pharmaceutical Sciences, University of Michigan, Ann Arbor, Michigan 48109-1065, United States

Abstract:

Tadalafil (TDF), used for the treatment of erectile dysfunction, is a practically water insoluble

drug. A cocrystal of TDF with malonic acid (MOA) was identified and characterized using a fleet

of solid-state characterization techniques including differential scanning calorimetry (DSC),

thermogravimetric analysis (TGA), Raman spectroscopy and powder X-ray diffraction (PXRD)

and single crystal X-ray diffraction technique. Cocrystal pH-solubility and supersaturation index

(SA) (defined as solubility of cocrystal/ solubility of drug) were determined and theoretical

models were applied. The dissolution behavior of the cocrystal and amorphous forms in the

presence and absence of HPMC was compared. The crystal structure analyses reveals the

formation of 1:1 ratio of TDF and MOA in asymmetric unit, wherein TDF molecule shows

persistent N–H…O hydrogen bonding interactions and catemeric chain of MOA forms hydrogen

bonds with lactam carbonyl group of TDF. MOA sublimes from the cocrystal lattice, followed by

the melting of TDF upon heating the cocrystal in the DSC. TDF-MOA showed higher solubility

than TDF over a wide pH range and it was dependent on the pH, with higher solubility at higher

pH values. Theoretical models based on Ksp have nicely predicted the cocrystal experimental

solubility-pH dependence. Supersaturation index for TDF-MOA cocrystal increased from

approximately 100 at pH 1 to about 48000 at pH 6.8. In the powder dissolution studies, cocrystal

and amorphous forms generated a supersaturation (Cmax/Sdrug) of 30 and 10, respectively and in

the presence of 0.1% w/v HPMC, both forms showed a supersaturation of 120 which was

maintained for at least 3 hours due to precipitation inhibition. Thus, SA based on equilibrium

studies can provide insights on the supersaturation level and the rate of cocrystal conversions in

the dissolution studies.

2

Key words: Tadalafil, cocrystal, crystal engineering, amorphous, pH-solubility, supersaturation index, dissolution.

1. Introduction

Active Pharmaceutical Ingredients (APIs) are often formulated into solid dosage forms

due to their favorable properties such as thermodynamic stability, purity, convenience and

excellent patient compliance. Different solid forms of an API such as polymorphs, salts, solvates,

amorphous form and cocrystals show different physicochemical properties, which can profoundly

affect the bioavailability of a resulting dosage form.1-3 physicochemical properties of a crystalline

form (single component or multicomponent) are inherently dependent upon the composition and

the packing of molecules in the crystal lattice.4-6 In this context, crystal engineering7 is an

important tool used as a design strategy to alter the crystalline lattice and improve the properties

(e.g. thermodynamic stability, solubility and mechanical etc.) of drug substances.8-10

Crystal engineering is the understanding of intermolecular interactions in the context of

crystal packing and the utilization of such understanding in design of new solids with desired

physical and chemical properties.11 Pharmaceutical cocrystals, often designed based on crystal

engineering principles, have been found to be effective in improving physical and technical

properties of clinical relevance.12, 13 In fact, a number of reports has revealed the potential of

cocrystals as supersaturating drug delivery systems.14, 15 For example, indomethacin-saccharin

(IND-SAC) cocrystals showed improved solubility and bioavailability of indomethacin as

compared to the crystalline form.16 The cocrystal eutectic point is an isothermal invariant point

where the solid cocrystal and drug are in equilibrium with solution and models have been

developed to predict cocrystal solubilities in pure solvent from measurement of transition

concentrations.17 The solubility advantage (or supersaturation index) of IND-SAC cocrystals was

estimated to be about 65 times over the crystalline form at pH 3 and the increase in cocrystal to

drug solubility was related to solvation effects.16 Furthermore, the aqueous solubility of

carbamazepine cocrystals was found to be 2-152 times greater than the solubility of

carbamazepine dihydrate form.16 Rodriguez-Hornedo et al. have derived mathematical models

based on cocrystal dissociation and ionization of the components that explain cocrystal solubility

dependence on the pH, pKa, cocrystals stoichiometry and coformer concentration.15, 17, 18 Since

solution conditions such as pH complexation determine the cocrystal solubility and stability,

cocrystals should be characterized not only by their solubility but also supersaturation behavior.15

3

Tadalafil (TDF, Figure 1) is a phosphodiesterase inhibitor,19 a pharmaceutical active ingredients,

under the brand name Cialis approved by Food and Drug Administration (FDA) in 2003 for the

treatment of erectile dysfunction (ED).20 TDF is a white crystalline powder, practically insoluble

in water21 and consider as a class II drug by Biopharmaceutical classification system.22 TDF is a

neutral drug with pKa of 15.17. TDF logP is 1.54 that reflects poor drug solubility in water.23

Many approaches have been tested and described in the literature to increase the dissolution rate

and bioavailability of TDF. The approaches reported include the use of (1) inclusion complexes,24

(2) solid dispersion,25-27 (3) amorphous hot-melt extrudate,28 (4) nanosuspension,29 and (5)

crystalline solvates forms.30 Although Weyna and co-workers have reported four cocrystals of

TDF, no dissolution and supersaturation performance was studied.31 Furthermore, cocrystals of

TDF and Malonic acid have not been identified. Malonic acid (MOA, Figure 1), is a

pharmaceutically accepted second-smallest aliphatic dicarboxylic acid.

The aim of the study was to identify cocrystals of tadalafil-malonic acid (TDF-MOA)

cocrystal and to perform comprehensive solid state and structural characterization. Another

objective was to determine pH-solubility dependence and provide insights on the solubility

advantage (or supersaturation index) offered by the cocrystals. It was also our aim to compare

supersaturation behavior of the cocrystal and amorphous forms in the powder dissolution studies

in the presence and absence of the precipitation inhibitors like HPMC. Finally, we aim to

understand how cocrystal supersaturation index (SA) determined under equilibrium conditions

relates to drug supersaturation levels during cocrystal dissolution.

N

N

HN

O

O

O

O

H3C

O

HO

O

OH

Figure 1. Molecular structures of Tadalafil (TDF) and coformer, malonic acid (MOA).

Hydrogen bond donors and acceptors are highlighted in red and blue respectively.

4

2. Materials and Methods

2.1 Materials

All chemicals and solvents (Purity > 99%) were purchased from Sigma Aldrich or Acros

Organics and were used without further purification. HPMC (Pharmacoat grade 606) was

obtained from Shin-Etsu chemical company (Japan) and all buffer salts for buffer solution

preparation were purchased from Sigma Aldrich (Stockholm, Sweden).

2.2 Methods

2.2.1 Screening for cocrystals of TDF

Non-stoichiometric slurry crystallization method was used for the screening of TDF

cocrystals. Based on the solubilities of TDF and various coformers (including,

nicotinamide, tartaric acid, citric acid, adipic acid, glutaric acid, malonic acid, succinic

acid, malic acid and vanillin), ethyl acetate and acetonitrile were used as solvents for

screening. Excess amount of TDF was added to the saturated solutions of coformers and

stirred at room temperature in a tightly caped glass vials. Solids were isolated after 24-48

h and subjected to PXRD and Raman spectroscopy or DSC. 2.2.2 Preparation and scale-up of TDF-MOA

A 7-mL glass vial was charged with 389 mg (1 mmol) of TDF, 104 mg (1 mmol) of

MOA. 3 mL of ethyl acetate was added to form slurry. The cake formation was observed,

indicating product formation. To make slurry, 3 mL of ethyl acetate was added in a glass

vial. The reaction was allowed to stir for a total of 5 hours, at which time the solids

present were isolated by vacuum filtration and air dried at room temperature to yield 419

mg (85%) of TDF-MOA cocrystal. Similar sizes of batches were prepared as per

experiment requirements.

2.2.3 Preparation of single crystals of TDF-MOA cocrystal

Approximately 30-60 mg of TDF-MOA cocrystal was used in the crystallization

experiments. For slow evaporation, the cocrystal was dissolved in enough acetonitrile (~2

mL) to produce an under saturated solution at room temperature. The solution was

collected by filtration in a 5-mL glass vial and let it stand to evaporate at low temperature

(~10-15oC). Colorless needle-shaped diffraction quality crystals of TDF-MOA (examine

under optical microscope) were obtained within 48 hours and isolated from the solution.

5

2.2.4 Preparation of amorphous TDF

Amorphous TDF was prepared by spray drying of 2% w/v solution of drug in acetone-

water (9:1, v/v), following the work by Wlodarski et al.32 Buchi mini spray dryer B-290

(Büchi, Switzerland) was used. An open system was used with nitrogen as a drying and

atomizing gas (open, suction mode). Spray drying process conditions were as follows:

inlet temperature 160 ˚C, pump speed 30 %, and outlet temperature 89 ˚C. The aspirator

was operated at 100%. The obtained amorphous powder was sieved (500 micron) and

analyzed by DSC and PXRD to confirm the amorphous state. The powders were stored at

-20 ˚C before further analysis

2.2.5 Solid state characterization

2.2.5.1 Raman spectroscopy

The Raman spectra were recorded on a Chromex Sentinel dispersive Raman unit equipped

with a 785nm, 70 mW excitation laser and a TE cooled CCD. Each spectrum is a result of

twenty co-added 20 second scans. The unit has continuous automatic calibration using an

internal standard. The data was collected by SentinelSoft data acquisition software and

processed in GRAMS AI.

2.2.5.2 Powder X-ray diffraction (PXRD)

PXRD patterns for the samples were collected using an Empyrean X-ray diffractometer

(PANalytical, The Netherlands) equipped with a PIXel3D detector and a monochromatic

Cu Kα radiation X-ray tube (1.54056 Å). The tube voltage and amperage were set at 45

kV and 40 mA, respectively. Samples were prepared for analysis by pressing a thin layer

of the sample onto a metal sample holder. Instrument calibration was performed using a

silicon reference standard. Each sample was scanned in the 2θ range of 5−40°, increasing

at a step size of 0.02° 2θ. The data were processed using HighScore Plus Version 3.0

software (PANalytical, The Netherlands)

2.2.5.3 Differential scanning calorimetry (DSC)

Thermal analyses of cocrystal samples were performed on a TA instruments DSC Q1000

(V9.9) The sample (1-3 mg) was placed into a non-hermetic aluminum DSC pan, and the

weight accurately recorded. The pan was crimped and the contents were heated (10 oC

6

min-1) under continuously purged dry nitrogen atmosphere (50 mL min-1) in the range of

25-320 oC. The instrument was equipped with a refrigerated cooling system. Indium was

used as the calibration standard. The data was collected in duplicate for each sample.

2.2.5.4 Thermogravimetric (TG) analysis

Thermogravimetric experiments were performed using a Thermal Advantage TGA Q5000

(V3.17) Instruments. Each sample (5-10 mg) was placed in a platinum pan (100 l) and inserted

into the TG furnace. The sample was heated with heating rate 10 oC min-1 in the rage of 25-350 oC, under a nitrogen purge gas (balance gas and sample gas was 10 and 90 oC min-1 respectively).

2.2.5.5 Variable temperature-powder X-ray diffraction (VT-PXRD)

High-temperature studies were carried out using an X-ray chamber HTK-10 Anton Paar. The

crystalline phase of the sample was determined by powder X-ray diffraction (PXRD) using a

PANalytical Empyrean diffractometer equipped with a Cu LFF HR X-ray tube and a PIXcel3D

detector. For the in situ high temperature measurement, the temperature was kept at 25 oC by

Anton Paar furnace with the heating rate of 50 oC/min. The diffraction patterns were recorded

every 50 °C interval. The diffraction patterns were recorded by 0.05° step scanning in the range

of 2θ angles from 5 to 30 grad. A platinum slice was used as a sample holder.

2.3 Single-crystal X-ray diffraction

The single-crystal X-ray diffraction data of the crystals were collected on a Bruker Nonius

Kappa CCD. The data set for compounds were collected at room temperature (293 K).

The crystals were quite stable and hence no extraordinary precautions were necessitated

for the smooth collection of intensity data. All the data sets were collected using Mo Kα

radiation (λ= 0.71073 Å), and crystal structures were solved by direct methods using SIR-

92 and refined by full-matrix least-squares refinement on F2 with anisotropic

displacement parameters for non-H atoms using SHELXL-2013. Hydrogen atoms were

placed on their expected chemical positions using the HFIX command. All the non-

hydrogen atoms were refined anisotropically. Structure solution, refinement, and

generation of publication materials were performed by using shelXLe, V623 software.33

All packing diagrams are generated using Diamond software.34 ORTEP diagram were

generated using OLEX2 software.35

7

2.4 Solubility and dissolution studies

2.4.1 Solubility of TDF

The equilibrium solubility of TDF was measured in a pH range of 1-7 and the pH of the

solution was adjusted by 1 M HCl or NaOH. An excess amount (approx. 200 mg) of

crystalline TDF was added to 5 mL solution of 0.1 N HCl and slurried in a water bath at

25 ˚C with orbital shaking (60 rpm.) After equilibration for 48-72 hours, the sample

solutions were filtered (using 0.2 μm PTFE filters) and analysed by HPLC (Agilent

Technologies, Stockholm). The pH at equilibrium was measured and the excess solids

were analysed using PXRD to confirm the solid form in equilibrium.

2.4.2 Cocrystal solubility and determination of eutectic concentrations of TDF and MOA

The eutectic point between cocrystal and drug was reached by simply adding 200 mg of

cocrystal powder (500 micron) in a 5 mL of solution at different pH. A 0.1 N HCL

solution was used for (pH 1, 2, and pH 5), acetate buffer solution for (pH 3 and 4) and

phosphate buffer solution for (pH 6 and 7). The solutions were subjected to orbital

shaking (60 rpm) for 48-72 hours at 25˚C and the pH at equilibrium was measured. The

concentration of the drug and the coformer at equilibrium at the eutectic point was

determined by HPLC. The eutectic point was confirmed by PXRD analysis of the residual

solid sample at equilibrium.

2.4.3 Dissolution experiments

Non-sink powder dissolution experiments were conducted to investigate dissolution and

supersaturation behaviour of amorphous and cocrystals of TDF. Excess amounts (approx.

1 mg/mL) of the crystalline TDF, amorphous TDF and equivalent amount of the cocrystal

powder were suspended in phosphate buffer solution (pH 6.8). Experiments were

conducted at 25 ˚C and 60 rpm orbital shaking. Aliquot samples were withdrawn at

different time points until 3 hours, filtered (0.2 μm PTFE filters), diluted and analysed by

HPLC. Dissolution of crystalline TDF, amorphous TDF and cocrystals were also

investigated in the presence of hydroxypropyl methyl cellulose (HPMC). A 0.1% w/v

HPMC was predissolved in the phosphate buffer solution (pH 6.8) and the dissolution

experiments were conducted under similar condition as before. At the end of dissolution

8

study, pHs of the solutions were measured and the separated solids were subjected to

PXRD analysis.

2.4.4 HPLC method

TDF and MOA were quantified using a quaternary pump HPLC unit (1290 series)

equipped with a UV detector (Agilent Technologies, Waldbronn), following methods.36, 37

A C-18 Nucleosil column (150×4.6 mm, 5 microns) (Sigma Aldrich, Sweden) was used

for both compounds. In the case of TDF, the mobile phase was 20 mM Na phosphate

buffer (pH 7) and Acetonitrile (30:70). Analysis was performed at a flow rate of 0.5

mL/min and detector wavelength 260 nm. Injection volume was 10μL. For the analysis of

MOA, mobile phase consisted of 25 mM potassium dihydrogen phosphate (pH 2.5) and

Methanol in 93:7 ratios. Flow rate, injection volume and the wavelength for UV detection

were 1 mL/min, 5 μL and 210 respectively.

3. Result and Discussion

3.1. Screening and identification of TDF-MOA cocrystal

A Cambridge Structural Database (CSD;38 ConQuest version 5.38, May 2017, Filters- only

organic molecules, number of hits 8) analysis revealed only six organic molecules of TDF;

including TDF39 (IQUMAI), two solvates30 (KAJKOY and KAJKUE) and three cocrystals31

(LASZUC, LATBAL and LATBEP). The crystal structure analysis of TDF reveals that TDF

molecules form supramolecular chains utilising hydrogen bond interactions present between the

indole group and the lactam carbonyl group of an adjacent TDF molecules. Due to the presence

of such hydrogen bond interactions, it is predisposed to form hydrogen bonds with a second

molecule (coformer) in the crystal lattice. Weyna et al. have utilized this knowledge in the design

of pharmaceutical cocrystals of TDF with methylparaben, propylparaben and hydrocinnamic

acid.31 In all TDF cocrystals, lactam carbonyl group, hydrogen bond acceptor of TDF is linked to

the hydrogen bond donor of coformers. Aakeröy et al.40 and Adsmond et al.41 engineered ternary

cocrystals based on observations of the hydrogen bond donor/acceptor pairings. Therefore,

supramolecular synthon history was used for cocrystal screening and included variety of

coformers that consist of hydrogen bond donor functionality in this study. Cocrystal screening

was conducted using slurry crystallization method, because it is a more efficient and considers

thermodynamic phase behaviour of different phases.8, 12 New pharmaceutical cocrystal of TDF

was found with coformer MOA as confirmed by initial solid-state characterization.

9

Figure 2. Raman spectra for (a) TDF, (b) MOA and (c) TDF-MOA cocrystal. The arrows and rectangles shows new or change in vibrational frequency in cocrystal form in comparison with cocrystal components.

The Raman spectrum of a new solid form was compared with the pure components

(Figure 2). The appearance of new vibrational peak at 1736 cm-1 and the disappearance shoulder

peak around 1600 cm-1 in cocrystal solid form indicate the formation of a new solid form (Figure

2). There are also additional peaks, as highlighted by arrows and rectangles, suggest the change

in intermolecular interactions in the cocrystal. PXRD is a rapid analytical technique used for

phase identification of crystalline materials. PXRD patterns for TDF, MOA, and TDF-MOA are

shown in Figure 3. The diffraction pattern for the new form was distinctly different from that of

TDF and MOA, indicating the formation of cocrystal. Thus, PXRD results corroborates with the

Raman spectra confirming the formation of tadalafil-malonic acid (TDF-MOA) cocrystal. Further

evidence of new cocrystal formation is presented in the following sections based on solid-state

techniques and single crystal X-ray diffraction.

10

Figure 3. Powder XRD patterns for (a) TDF, (b) TDF-MOA cocrystal and (c) MOA. The asterisks denote new diffraction peaks in cocrystal form in comparison with cocrystal components.

3.2 Crystal structure description

The crystal structure was elucidated using single crystal X-ray diffraction. The obtained

data revealed that TDF-MOA crystallizes in the orthorhombic space group P212121‡ with

one molecule each of TDF and MOA in the asymmetric unit as shown in Figure 4. MOA

is the coformer possessing carboxylic acid groups capable of forming strong hydrogen

bonds with or without proton transfer. The structural analysis found that the carbonyl

group of malonic acid has shorter bond length (C24–O7 = 1.22 Å and C25–O6 = 1.19 Å)

and C–OH group of malonic acid has larger bond length (C24–O8 = 1.33 Å and C25–O5

= 1.34 Å. Thus, malonic acids in the cocrystal exist as a neutral moiety (no proton was

transfer occurred), which is evident from the carboxylic acid bond length.

11

Figure 4. ORTEP of asymmetric unit in the crystal lattice of TDF-MOA cocrystal.

The crystal packing in cocrystal shows that adjacent TDF molecules are hydrogen bonded

via N–H…O=C interactions [N3–H3…O1; N…O = 2.971 Å, N–H…O = 158.59o] of a neighbouring

molecule (the indole group and the lactam carbonyl group) as shown in Figure 5a. The CSD38

analysis of different TDF crystal forms shows that the adjacent TDF molecule shows persistent

N–H…O hydrogen bonding interactions in all the reported TDF crystal form.30, 31, 39 The persistent

hydrogen bonding motif (bond distance and angle) present between the indole and the lactam

carbonyl group an observed in all TDF crystal forms are tabulated in Table 1. Table 1. Characteristics of important hydrogen bonds observed in different TDF crystal forms.

Crystal structure with CSD code N-H…O=C interactionsa N…O (Å) H…O (Å) N–H…O (o) TDF (IQUMAI) 2.985 2.09 167.1 TDF-MPB (LASZUC) 2.981 2.12 164.4 TDF-PPB (LATBAL) 2.965 2.10 168.3 TDF-HCA (LATBEP) 2.918 2.05 168.3 TDF-ACE (KAJKOY) 2.947 2.11 165.7 TDF-MEK (KAJKUE) 2.957 2.11 166.1 TDF-MOA (1581033) 2.971 2.15 158.6

aThree columns for each structure represent N…O /H…O bond length in angstroms and bond angle in degrees respectively.

The supramolecular chains of TDF are linked to the MOA molecules through C=O…H–O

hydrogen bonds [C=O2…H8a–O8; O…O = 2.605 Å, O–H…O = 158.36 o] as shown in Figure 5b.

12

Further, MOA forms the catemeric chains in the crystal lattice along a 21 screw axis

perpendicular to the b-axis through O–H…O hydrogen bonds [O5–H6e…O7; O…O = 2.718 Å, O-

H…O = 155.60o] (Figure 5b). TDF molecule forms double layered structures (Figure 5a) through

N–H…O=C interactions and this double layers are separated by catemeric chains of MOA (Figure

5b). In order to check the homogeneity and crystalline phase purity of prepared TDF-MOA

cocrystal by reaction crystallization, the experimental and simulated PXRD pattern (generated

from the single crystal structure of TDF-MOA using Mercury42) were compared. The

experimental PXRD did match the simulated PXRD (standard reference) pattern from the single-

crystal structure (Figure S1) confirm the homogeneity and purity of TDF-MOA crystalline phase.

In addition, single endothermic transition observed in DSC curve further attest the single

crystalline phase present in the cocrystal (Figure 6a, discussed in details in 3.3 Section).

Figure 5. (a) A catamer-like chain via N-H…O hydrogen bonds along the b-axis. (b) The

catemeric chains are separated by malonic acids zig-zig chain via O-H…O hydrogen bonds along

the b-axis. To better visualize/understand the potential intermolecular interactions of cocrystal,

1,3-benzodioxole group was removed.

3.3 Thermal behaviour of TDF-MOA cocrystal

The thermal behaviour of TDF-MOA cocrystal in relation to the TDF and MOA was

carried out using DSC and TGA. The DSC thermogram for TDF, MOA and TDF-MOA

cocrystal are presented in Figure 6a. TDF showed a single endothermic transition with

(a) (b)

13

Tonset = 304.2 oC. MOA shows first endothermic peak at 96.2 oC due to solid phase

transition; second endothermic transition at Tonset = 134.4 oC attributed to melting; and a

third endothermic peak at 178.8 oC corresponding to the thermal decomposition. The

overlaid DSC and TGA thermograms of MOA are presented in Figure S2. Thermal events

observed for MOA are in agreement with the reported thermal behaviour.43 The DSC

thermogram for TDF-MOA cocrystal showed a first endothermic transition at Tonset (∆Hf)

192.7 oC 2.0 (139.0 J g-1 9.0) and second endothermic peak at Tonset (∆Hf) 280.3 oC 2.7

(55.71 J g-1 5.4). The DSC thermogram of the cocrystal was distinctly different from

individual components as observed in Figure 6a. Thermal behaviour of the TDF-MOA

cocrystal was further investigated by TGA. Interestingly, TGA of TDF-MOA shows first

weight loss of around 25% at ~195 oC where the major endothermic event (Tonset = 192.7 oC) occurs in the DSC, which suggests a possible gas released event (Figure 6b). The

second major weight loss at >310 oC was possibly due to the degradation of the material

(Figure 6b). To get more insights on the first weight loss, TDF-MOA cocrystal was

analysed by capillary melting point apparatus. Release of gases from cocrystal powder

material was observed in the temperature range 185-200 oC. Subsequent heating of the

powder resulted in a complete melting at ~300 oC.

Figure 6. (a) DSC thermograms for tadalafil, TDF; malonic acid (MOA) and TDF-MOA cocrystal. (b) DSC (blue line) and TG (black line) overlaid profile of the TDF-MOA cocrystal.

In a separate experiment, the cocrystal powder was heated in a capillary to 210 oC, (i.e.

until after the gases are released but essentially before the melting event) and the powder

sample was isolated at room temperature and subjected to Raman analysis. As observed in

(a) (b)

14

Figure S3, the Raman spectra were found to be identical to that of TDF. It was reported

that the subsequent heating after the melting of MOA (at around 137 oC) results in

endothermic event at Tonset = 190.7 oC, corresponding to the evaporation of acetic acid

and carbon dioxide gases as confirmed by IR spectroscopy.43 Thus, the capillary melting

experiments and DSC/TGA data and its analyses suggest the sublimation phenomenon

(i.e. solid to gas transformation) of MOA was observed in the cocrystal at 190 oC. We

have also collected further evidence of sublimation phenomenon using variable

temperature powder X-ray diffraction (VT-PXRD). The characteristic diffraction peaks of

the cocrystal were observed until 180 oC (Figure 7) later it convert into crystalline, in line

with DSC and TGA observations. Thus, based on the solid-state analysis, we conclude

that the TDF-MOA cocrystal undergoes sublimation of MOA from the cocrystal lattice, in

the form of gases, followed by the melting of TDF. To the best of our knowledge, this is

first example of a cocrystal wherein the coformer sublimates from the crystal lattice

leaving behind the drug which subsequently melts. This is another intriguing example

where the cocrystallization can impart drastic change in thermal behaviour of the drug or a

conformer (i.e. patent compounds).

Figure 7. Variable temperature powder X-ray pattern for the TDF-MOA cocrystal. The

characteristic peaks of cocrystal and TDF are highlighted using rectangular boxes.

15

3.4 TDF-MOA cocrystal solubility and

Understanding cocrystal solubility-pH dependence helps to guide formulation

development of cocrystals as well as to determine supersaturation index and

conversion/transformation propensity of cocrystals. Furthermore, knowledge of how

cocrystal solubility-pH is affected by cocrystal components provides tools to engineer

cocrystals with customized solubility behaviour. In fact, equations that describe cocrystal

solubility in terms of solubility product, cocrystal component ionization constants, and

solution pH have been derived for various cocrystals with acidic, basic, amphoteric, and

zwitterionic components.15 TDF-MOA (1:1) cocrystal consists of non-ionic drug, TDF

and one diprotic acid MOA, with two pKa values 2.83 and 5.69. While detailed

derivations can be found elsewhere,15 the TDF-MOA cocrystal solubility dependence on

is given by,

(1)

where is the cocrystal solubility product, and are the ionization constants of

MOA. was calculated as the product of the concentrations of the nonionized

constituents according to the following equation;

(2)

Where and are eutectic concentrations of TDF and MOA respectively.

Table 2. Cocrystal and intrinsic solubility of the drug, STDF.

Cocrystal (M2) STDF (M) SCC/STDF (pH 1-3)

TDF-MOA

cocrystal (3.8 ± 1.4 ) × 10-07

(4.6 ± 0.2) × 10-6 134-210

S refers to solubility under nonionizing conditions ± represents standard deviation

16

TDF-MOA was found to convert back into TDF at pH 1-7.4 and the solubility was

measured at eutectic point, where the solution phase is in equilibrium with both cocrystal

and drug solid phases.17 Figure S4 shows presence of cocrystal and drug phases at the

eutectic points at different pH. Cocrystal was determined from equation 2.

Table 2 summarizes the cocrystal and intrinsic solubility of the drug, STDF. was

used to generate the cocrystal solubility dependence on pH. Figure 8 shows that the TDF-

MOA solubility is higher than TDF over a wide pH range. For example, even at its lowest

solubility, cocrystal appears to be over 100 times more soluble than the drug (Figure 8 and

Table 2). These trends are consistent with coformer solubility and ionization properties, in

line with the results reported for other cocrystals.16, 17

Experimental cocrystal solubilities could only be measured in the pH range (pH 1-3) as

the cofomer controlled/lowered the pH of the solution. Mathematical model (eq 1), based

on and values, predicts the exponential increase in cocrystal solubility with pH and

these theoretical values are in excellent agreement with the experimentally measured

solubilities (Figure 8). Thus, it is possible to predict the cocrystal solubility-pH

dependence from knowledge of Ksp, at pH values where solubility was not experimentally

measured and to gain insight of the supersaturation behavior of cocrystals at

physiologically relevant pH. It is clear from the shape of solubility-pH curve that the

TDF-MOA cocrystal imparts pH dependent solubility to TDF, demonstrating the potential

of cocrystal technology in engineering solubility behavior of nonionizing drugs as a

function of pH. Previously, cocrystals of saccharin were shown to not only enhance

aqueous solubility but also impart pH sensitivity to nonionizing drugs like

carbamazepine.16 Furthermore, TDF eutectic concentrations are in agreement with drug

solubility values (STDF) values (Figure 8) suggesting that there is no drug complexation

with coformer in solution.

17

Figure 8. Solubility-pH dependence of 1:1 TDF-MOA cocrystal and TDF. Experimentally measured cocrystal and drug solubilities are represented by (red ○) and (blue □) respectively. concentrations are shown in (blue ○). Lines represent theoretical solubilities of cocrystal (red dashed line) and drug (blue line). Solubility-pH dependence was calculated according to eq 1 with value listed in Table 2 and pKa values of 2.83 and 5.69 for MOA. STDF (line) was determined as the average of experimental TDF solubility points (blue squares).

3.5 Supersaturation index (SA) of TDF-MOA cocrystals

The enhanced solubility of drug cocrystals has been reported for a number of drugs.8, 44, 45 The

solubility advantage or supersaturation index (SA), defined as Scc/Sdrug, represents the driving

force for drug precipitation and provides a metric for the risk of cocrystal conversions. It also

provides guidance for the selection of pH and formulation additives to prevent drug precipitation.

Recently, Rodriguez-Hornedo el al. have introduced a simple method to tailor cocrystal SA by

the rational selection of drug solubilizing agents and pH.46-48

Supersaturation index points were calculated from experimental cocrystal solubility and eutectic

drug concentrations. Theoretical SA is extrapolated according to

18

(3)

Figure 9 shows the SA dependence on pH for TDF-MOA cocrystal. Evidently, theoretical

and experimentally measured values are in good agreement. Interestingly, SA index for

TDF-MOA cocrystal increases from approximately 100 at pH 1 to about 48000 at pH 6.8

(Figure 9). SA values are orders of magnitude higher than experimentally measured

supersaturation generated by amorphous or cocrystal forms due to onset of nucleation at a

lower supersaturation value. SA-pH dependence graph provides a means to assess slow

and fast cocrystal conversions to less soluble drug. High SA values are expected to lead to

fast cocrystal conversions, fast drug nucleation, and low observed supersaturation.

Thus, the maximum daily TDF dose of 20 mg (dose number 26, TDF solubility 8× 10-07

M ( 0.3 μg/mL) and volume of solution 250 mL) is well below the cocrystal solubility

under physiological conditions (pH 1-7.4). In studying solubility advantage of cocrystals,

Good and Rodriguez-Hornedo have shown that the cocrystal solubility is directly

proportional to the solubility of constituent reactants for a wide range of drugs (including

acidic, basic, zwitterionic and neutral).14 Later, Velaga et al have revealed increase in the

coformer concentration at the eutectic point for carbamazepine-saccharin cocrystals and

concluded that the cocrystal aqueous solubility advantage was a result of decrease in the

solvation energy (log γ).16

19

Figure 9. Cocrystal supersaturation Index (cocrystal solubility advantage) as a function of pH for TDF-MOA cocrystal. Symbols represent the calculated Scocrystal/Sdrug values from solubility measurements and line represents theoretical dependence on pH generated from eq 3 with values listed in Table 2 and pKa values of 2.83 and 5.69 for MOA.

3.6 Powder dissolution and supersaturation behaviour of TDF solid forms

Figure 10 shows non-sink dissolution profiles of TDF cocrystal and amorphous solid in

the presence and absence of precipitation inhibitor HPMC at pH 6.8. In the absence of

HPMC, cocrystals and amorphous forms generate faster dissolution and supersaturation

with respect to TDF, followed by precipitation (Figure 10). Whilst in the presence of 0.1%

w/v HPMC, cocrystal and amorphous forms achieved much higher concentrations than the

drug and the supersaturation was sustained for 3 hours (Figure 10).

TDF crystalline form reached a plateau that corresponds to the equilibrium solubility of

the drug 9.23× 10-07 M ( 0.36 μg/mL) in phosphate buffer pH 6.8. The spray dried TDF

(amorphous form) and cocrystals dissolved very quickly and reached maximum solution

concentrations (Cmax) of 9.02 × 10-06 M and 2.61 × 10-05 M respectively (3.51 and

10.12 μg/mL respectively), in less than 5 min. Thereafter the concentrations decreased

because of precipitation or conversion of cocrystal to TDF as confirmed by PXRD

20

analysis (supplementary information). The dissolution behavior of TDF-MOA resembles

many other cocrystals in the literature.44 For example, the ezetimibe–L-proline cocrystal

showed rapid dissolution and conversion to ezetimibe anhydrate in phosphate buffer (pH

6.5).8 The maximum concentration is expressed in terms of supersaturation (Cmax/Sdrug),

which represents the concentration above which precipitation is faster than the dissolution

of the cocrystal. The cocrystal and amorphous forms generated a supersaturation of 30

and 10, respectively (Figure 10).

TDF-MOA cocrystal in the presence of 0.1% w/v HPMC showed Cmax of 9.5 × 10-05

M ( 37.25 μg/ml), which corresponds to a supersaturation of about 120. Amorphous form

also achieved similar concentrations. For both cocrystal and amorphous phase, the

supersaturation was maintained much longer (3 hours) in the presence of HPMC i.e.,

suggesting slower conversion of amorphous or cocrystal form to drug. HPMC at a

concentration of 0.1% w/v was found to provide maximum precipitation inhibition from

solution supersaturated at 29.25 These findings suggest that cocrystals may increase drug

bioavailability under similar conditions.

Cmax during cocrystal dissolution reached a supersaturation of about 30 at SA of 10000

(Figure 10). At this SA, HPMC was effective in slowing drug nucleation, reaching

supersaturation values of 120. Similarly, SA was shown to increase with increasing the pH

for cocrystals of ketoconazole, and a faster precipitation of the drug was observed.49

Therefore, SA appears to be an indicator of the rate of cocrystal conversions.

21

Figure 10: Dissolution profiles of different solid forms of TDF in phosphate buffer (pH

6.8±0.4); Crystalline TDF (Blue), Amorphous TDF (Green), TDF-MOA cocrystal (Red)

and phosphate buffer containing 0.1% w/v of HPMC; Amorphous TDF (Black) and TDF-

MOA cocrystal (Purple) in Error bars represent standard deviations obtained from

triplicate measurements. The physical purity of cocrystals and amorphous forms was

confirmed by PXRD and DSC (Figure S4).

Conclusion

Cocrystal of Tadalafil with malonic acid was identified using crystal engineering strategy.

TDF-MOA cocrystal material was successfully scaled-up to few grams using solution

crystallization method and thoroughly characterized. TDF-MOA cocrystal crystallized in

chiral space group P212121, with one molecule of each TDF and MOA in the asymmetric

unit. In the TDF-MOA crystal structure, adjacent TDF molecule shows persistent N–H…O

hydrogen bonding interactions as observed in other TDF crystal forms. The catemeric

chain of MOA forms hydrogen bonds with lactam carbonyl group of TDF. Interestingly,

22

TDF-MOA cocrystal does not show classical melting point rather MOA sublimes from the

cocrystal lattice.

In the aqueous media, TDF-MOA cocrystal solubility was higher than TDF over a

wide range of pH. Solubility-pH dependence of the cocrystal and SA can be calculated

using cocrystal Ksp, pKa(s) and drug intrinsic solubility. Theoretical models nicely

predicted experimental pH solubility and SA results. Supersaturation index of TDF-MOA

cocrystal increased by order of magnitude with the increase in pH, reaching as high as

48000 at pH6.8. In powder dissolution studies, cocrystal and amorphous form generated a

supersaturation (Cmax/Sdrug) of 30 and 10, respectively. The addition of 0.1% w/v HPMC

to dissolution media increased supersaturation to 120 by slowing down nucleation and/or

precipitation kinetics of the drug from cocrystal and amorphous form. Therefore, the

supersaturation during cocrystal dissolution may not reach SA at such a high SA. Thus,

SA calculated from the cocrystal and drug solubility equations represents the driving

force for nucleation and provides insights on the cocrystal dissolution-precipitation

behavior of cocrystal.

Ackowledgements.

MRS grateful for the financial support from Kempe foundation for postdoctoral grant and The Knut & Alice Wallenberg foundations, Sweden. Professor Sung-Joo Hwang and his team at Yonsei University, South Korea acknowledged for preliminary screening experiments

‡ Crystallographic data and structure refinement parameters of TDF-MOA cocrystal.

empirical formula = C25H23N3O8, formula weight = 493.46, orthorhombic, space group =

P212121, a = 12.133(4) Å, b = 7.782(3) Å, c = 23.589(12) Å, V= 2227.25(16) Å3, Z = 4, Dcalc=

1.472 g cm-3, T= 293(2) K, μ = 0.111 mm-1, F(000) = 1032, number of reflection used = 1585,

number of parameters = 370, GOF on F2 = 1.118; R1[I>2σ(I)] =0.0777, wR2= 0.1825, CCDC

number = 1581033

23

References

1. Childs, S. L.; Kandi, P.; Lingireddy, S. R. Mol. Pharm. 2013, 8, 3112-3127.

2. Kobayashi, Y.; Ito, S.; Itai, S.; Yamamoto, K. Int. J. Pharm. 2000, 2, 137-146.

3. Jung, M. S.; Kim, J. S.; Kim, M. S.; Alhalaweh, A.; Cho, W.; Hwang, S. J.; Velaga, S. P. J.

Pharm. Pharmacol. 2010, 11, 1560-1568.

4. Shimpi, M. R.; Velaga, S. P.; Shah, F. U.; Antzutkin, O. N. Cryst. Growth Des. 2017, 4, 1729-

1734.

5. Srivastava, K.; Shimpi, M. R.; Srivastava, A.; Tandon, P.; Sinha, K.; Velaga, S. P. RSC Adv.

2016, 12, 10024-10037.

6. Pandey, J.; Prajapati, P.; Shimpi, M. R.; Tandon, P.; Velaga, S. P.; Srivastava, A.; Sinha, K.

RSC Adv. 2016, 78, 74135-74154.

7. Desiraju, G. Current Opinion in Solid State & Materials Science 1997, 4, 451-454.

8. Shimpi, M. R.; Childs, S. L.; Bostroem, D.; Velaga, S. P. CrystEngComm 2014, 38, 8984-

8993.

9. Ahmed, H.; Shimpi, M. R.; Velaga, S. P. Drug Dev. Ind. Pharm. 2017, 1, 89-97.

10. Jones, W.; Motherwell, S.; Trask, A. V. MRS Bull 2006, 11, 875-879.

11. Desiraju, G. R. Crystal Engineering. The Design of Organic Solids; Elsevier: Amsterdam,

1989; .

12. Malamatari, M.; Ross, S. A.; Douroumis, D.; Velaga, S. P. Adv. Drug Delivery Rev. 2017,

162-177.

13. Bolla, G.; Nangia, A. Chem. Commun. (Cambridge, U. K. ) 2016, 54, 8342-8360.

14. Good, D. J.; Rodriguez-Hornedo, N. Cryst. Growth Des. 2009, 5, 2252-2264.

15. Bethune, S. J.; Huang, N.; Jayasankar, A.; Rodriguez-Hornedo, N. Cryst. Growth Des. 2009,

9, 3976-3988.

16. Alhalaweh, A.; Roy, L.; Rodriguez-Hornedo, N.; Velaga, S. P. Mol. Pharmaceutics 2012, 9,

2605-2612.

17. Good, D. J.; Rodriguez-Hornedo, N. Cryst. Growth Des. 2010, 3, 1028-1032.

18. Huang, N.; Rodriguez-Hornedo, N. J. Pharm. Sci. 2011, 12, 5219-5234.

19. Laties, A. M. Drug Saf. 2009, 1, 1-18.

20. Gresser, U.; Gleiter, C. H. Eur. J. Med. Res. 2002, 10, 435-446.

21. El-Badry, M.; Haq, N.; Fetih, G.; Shakeel, F. J. Chem. Eng. Data 2014, 3, 839-843.

24

22. Benet, L. Z.; Broccatelli, F.; Oprea, T. I. Aaps J. 2011, 4, 519-547.

23. Krupa, A.; Majda, D.; Mozgawa, W.; Szlek, J.; Jachowicz, R. AAPS PharmSciTech 2017, 4,

1318-1331.

24. Badr-Eldin, S. M.; Elkheshen, S. A.; Ghorab, M. M. Eur. J. Pharm. Biopharm. 2008, 3, 819-

827.

25. Christfort, J. F.; Plum, J.; Madsen, C. M.; Nielsen, L. H.; Sandau, M.; Andersen, K.;

Mullertz, A.; Rades, T. Mol. Pharmaceutics 2017, Ahead of Print.

26. Choi, J.; Lee, S.; Jang, W. S.; Byeon, J. C.; Park, J. Eur. J. Pharm. Sci. 2017, 152-158.

27. Choi, J.; Park, J. Eur. J. Pharm. Sci. 2017, 269-276.

28. Krupa, A.; Cantin, O.; Strach, B.; Wyska, E.; Tabor, Z.; Siepmann, J.; Wrobel, A.; Jachowicz,

R. Int. J. Pharm. (Amsterdam, Neth. ) 2017, 1-2, 498-510.

29. Obeidat, W. M.; Sallam, A. A. AAPS PharmSciTech 2014, 2, 364-374.

30. Miclaus, M. O.; Kacso, I. E.; Martin, F. A.; David, L.; Pop, M. M.; Filip, C.; Filip, X. J.

Pharm. Sci. 2015, 11, 3782-3788.

31. Weyna, D. R.; Cheney, M. L.; Shan, N.; Hanna, M.; Wojtas, L.; Zaworotko, M. J.

CrystEngComm 2012, 7, 2377-2380.

32. Wlodarski, K.; Sawicki, W.; Paluch, K. J.; Tajber, L.; Grembecka, M.; Hawelek, L.;

Wojnarowska, Z.; Grzybowska, K.; Talik, E.; Paluch, M. Eur. J. Pharm. Sci. 2014, 132-140.

33. Hubschle, C. B.; Sheldrick, G. M.; Dittrich, B. J. Appl. Crystallogr. 2011, Pt 6, 1281-1284.

34. Crystal-Impact, Brandenburg & Putz, Bonn, 2008.

35. Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.; Puschmann, H. J. Appl.

Crystallogr. 2009, 2, 339-341.

36. Prasanna Reddy, B.; Amarnadh Reddy, K.; Reddy, M. S. Res. Pharm. Biotechnol. 2010, 1,

001-006.

37. Cawthray, G. R. J. Chromatogr. A 2003, 1-2, 233-240.

38. Allen, F. H. Acta Crystallogr. B 2002, Pt 3 Pt 1, 380-388.

39. Daugan, A.; Grondin, P.; Ruault, C.; Le Monnier de Gouville, A.; Coste, H.; Linget, J. M.;

Kirilovsky, J.; Hyafil, F.; Labaudiniere, R. J. Med. Chem. 2003, 21, 4533-4542.

40. Aakeroy, C. B.; Beatty, A. M.; Helfrich, B. A. Angew. Chem. , Int. Ed. 2001, 17, 3240-3242.

41. Adsmond, D. A.; Sinha, A. S.; Khandavilli, U. B. R.; Maguire, A. R.; Lawrence, S. E. Cryst.

Growth Des. 2016, 1, 59-69.

25

42. Cambridge Crystallographic Data Centre, UK Mercury 2017.

43. Caires, F. J.; Lima, L. S.; Carvalho, C. T.; Giagio, R. J.; Ionashiro, M. Thermochim. Acta

2010, 1-2, 35-40.

44. Babu, N. J.; Nangia, A. Cryst. Growth Des. 2011, 7, 2662-2679.

45. Basavoju, S.; Bostroem, D.; Velaga, S. P. Cryst. Growth Des. 2006, 12, 2699-2708.

46. Lipert, M. P.; Rodriguez-Hornedo, N. Mol. Pharmaceutics 2015, 10, 3535-3546.

47. Maheshwari, C.; Andre, V.; Reddy, S.; Roy, L.; Duarte, T.; Rodriguez-Hornedo, N.

CrystEngComm 2012, 14, 4801-4811.

48. Kuminek, G.; Rodriguez-Hornedo, N.; Siedler, S.; Rocha, H. V.; Cuffini, S. L.; Cardoso, S.

G. Chem. Commun. (Camb) 2016, 34, 5832-5835.

49. Chen, Y.; Rodriguez-Hornedo, N. Cryst. Growth Des. 2018, Ahead of Print.

26

For Table of Contents Use Only

Cocrystal of tadalafil: Physicochemical characterization,

pH-solubility and supersaturation studies Manishkumar R. Shimpi1,2, Amani Alhayali1, Katie L. Cavanagh3, Naír Rodríguez-

Hornedo3, Sitaram P. Velaga1, *

1Pharmaceutical and Biomaterials group, Department of Health Sciences 2Chemistry of Interfaces, Luleå University of Technology, SE-97187, Luleå, Sweden.

3Department of Pharmaceutical Sciences, University of Michigan, Ann Arbor, Michigan 48109-1065, United States

An important goal of pharmaceutical solid-state development is to increase drug solubility while maintaining a stable form. Cocrystal of tadalafil with malonic acid was identified using crystal engineering strategy. Interestingly, TDF-MOA cocrystal does not show classical melting point. The studies demonstrate maximum achievable supersaturation is dictated by the aqueous solubility of the cocrystal form of the drug.

Cocrystal of tadalafil: Physicochemical characterization,

pH-solubility and supersaturation studies Manishkumar R. Shimpi1,2, Amani Alhayali1, Katie L. Cavanagh3, Naír Rodríguez-

Hornedo3, Sitaram P. Velaga1, *

1Pharmaceutical and Biomaterials group, Department of Health Sciences 2Chemistry of Interfaces, Luleå University of Technology, SE-97187, Luleå, Sweden.

3Department of Pharmaceutical Sciences, University of Michigan, Ann Arbor, Michigan 48109-1065, United States

.

Figure S1. Experimental and simulated (obtained from single crystal data) powder X-ray diffraction data for TDF-MOA cocrystal.

Figure S2. DSC and TGA profile for Malonic acid.

Figure S3. Raman spectrum for TDF shown in black line. The blue line spectrum is obtained from the powder sample isolated when cocrystal was heated till 210 oC.

Figure S4. PXRD of Amorphous-TDF prepared by spray drying method. PXRD of powder materials isolated from solution of pH 1 and 3 was compare with PXRD of TDF and TDF-MOA cocrystal. The arrow indicates peaks corresponding to the cocrystal present in powder materials isolated from solution of pH 1 and 3.

Paper IV

1

Silodosin oral films: Development, physico-mechanical properties and in vitro

dissolution studies in simulated saliva

Amani Alhayalia*, Parameswara Rao Vuddandaa, Sitaram Velagaa

Pharmaceutical and Biomaterial Research Group, Division of Medical Sciences, Department

of Health Sciences, Luleå University of Technology, Luleå, Sweden

*Corresponding: author: Amani Alhayali

Email: amani.al-hayali@ltu.se

2

Abstract

Sublingual film dosage forms for drugs used for fast symptomatic treatment have promise

because they allow a rapid onset of action. The aim of this study was to prepare films of

silodosin intended for sublingual administration for the symptomatic treatment of benign

prostatic hyperplasia in men. Hydroxypropyl methylcellulose (HPMC) or hydroxypropyl

methylcellulose acetate succinate (HPMC-AS) were used as film-forming polymers. The

effects of the polymers and the surfactant tocopherol polyethylene glycol succinate (TPGS)

on the physico-mechanical properties and dissolution behavior of the films in simulated saliva

were investigated. The eight silodosin oral films developed (F1–F8) contained 8mg silodosin

per 6 cm2 film and HPMC or HPMC-AS in drug:polymer ratios of 1:5 or 1:3, while four also

contained TPGS (0.5 % w/w). The films were characterized using DSC, TGA, SEM, and

PXRD and the mechanical properties were investigated by measuring tensile strength,

elongation at break and Young’s modulus. The mechanical properties of the films were

dependent on the ratio of polymer used. The in vitro dissolution and drug release studies

indicated that HPMC-AS films disintegrated more quickly than HPMC films. Silodosin was

shown to be dispersed within the polymers. Despite silodosin being submicronized in the

HPMC films, the dissolution and drug release rate (time for 80% release) from HPMC films

was significantly faster than from HPMC-AS films. TPGS increased the drug release rate to a

greater extent with HPMC than with HPMC-AS. The degree of saturation of formulation F4

was >1, which shows potential for improving oral absorption of silodosin.

Keywords:

Silodosin, sublingual oral films, HPMC, HPMC-AS, TPGS, simulated saliva

3

1. Introduction

Conventional oral dosage forms, such as tablets or capsules, have challenges related to

dissolution, absorption and poor bioavailability for some drugs and can also be associated

with problems related to patient compliance {{ Singh, Harmanpreet 2015; Singh, Baljeet

2016; Sharma, Rahul 2014 }}. Alternative drug delivery systems such as oromucosal

formulations can be used to overcome these drawbacks {{Kalia, Vani 2016 }}. Oromucosal

formulations such as oral films have recently been receiving attention from the

pharmaceutical industry because of their unique advantages {{Mashru, RC 2005; Slavkova,

Marta 2015}}. For instance, oral films are easy to administer, do not require chewing or

intake of water, and disintegrate and dissolve rapidly to release the drug when placed in the

oral cavity {{De Caro, V 2012; Campisi, Giuseppina 2010 }}. This has the potential to

improve patient compliance, mainly for pediatric and geriatric patients but also for others with

mental disorders, dysphagia or emesis {{ Koland, M 2010; Mashru, RC 2005 }}. Drugs

formulated as oral films intended for sublingual administration will be directly and rapidly

absorbed into the systemic circulation without passing through the gastrointestinal tract (GIT),

thereby bypassing first-pass metabolism in the liver {{ Preis, Maren 2013; Hanif,

Muhammad 2015}}. The relatively extensive vascularity and high permeability of the

sublingual mucosa and membranes can facilitate rapid absorption of the formulated drug and

instant bioavailability {{Zhang, Hao 2002; Madhav, NV Satheesh 2009; Mashru, RC 2005;

Allam, A. 2016}}.

Oral films can be prepared by various methods, including solvent casting, hot-melt extrusion,

electrospinning, freeze drying and ink-jet printing {{Repka, Michael A 2003; Boateng,

Joshua S 2010; El-Setouhy, Doaa Ahmed 2010; Vuddanda 2018}}. The solvent-casting

method is appropriate and feasible for manufacturing on an industrial scale. Usually, oral

films contain polymers, plasticizers, drug, surfactants and taste-masking agents (sweeteners

and flavors) as required. The film-forming ability and water solubility are the main

considerations for selecting the polymer. The polymer to plasticizer ratio is also crucial, as

this affects the physico-mechanical stability of the final product, and consideration of this

attribute is thus required in the design and development process for oral film formulations.

Garsuch and Breitkreutz have reported that the cellulose-derived polymer hydroxypropyl

methylcellulose (HPMC) forms films well and has better mechanical properties than other

tested excipients{{Garsuch 2010}}. In another study, Visser et al. reported the optimal

physico-mechanical properties of films prepared with combined HPMC and polyol

4

plasticizers {{Visser 2015}}. Oral films can be considered as solid dispersions and the

systematic determination of the drug’s solubility in the film-forming excipients

(polymers/surfactants) using the appropriate techniques is an important development step.

Cellulose-based polymers such as HPMC and hydroxypropyl methylcellulose acetate

succinate (HPMC-AS) help to stabilize the physical structure of the drug, preventing it from

recrystallizing, and increasing its supersaturation during dissolution {{Curatolo 2009}}. It

could thus be interesting to evaluate the potential of these polymers in the preparation of oral

films intended for sublingual drug absorption. The oral cavity has a smaller surface area and

less dissolution medium (i.e. saliva) than the GIT and the polymers could provide a vital boost

to the dissolution and subsequent absorption of drugs administered as a film {{Patel,

Viralkumar F 2011}}. Furthermore, the film-forming properties and potential of HPMC and

HPMC-AS in film drug delivery have been extensively investigated.

Surfactants are also an important excipient in the preparation of drug-carrying films,

particularly for poorly water-soluble drugs. It is well known that surfactants can facilitate

wettability and enhance the dissolution rate of poorly water-soluble drugs. For example,

Vuddanda et al. have reported that the addition of surfactant enhanced the dissolution of

tadalafil nanocrystal-loaded oral films {{Vuddanda 2017}}.

Benign prostatic hyperplasia (BPH) is an enlargement of the prostate gland caused by the

proliferation of prostatic stromal cells. It’s an important cause of lower urinary tract

symptoms in men such as frequency, urgency, nocturia, and hesitancy. BPH is a common

problem among men after the age of 40 years {{Kapoor, Anil 2012; Rossi, M. 2010}}.

Silodosin is a selective α1A adrenoceptor blocker that is a safe, effective treatment for the

relief of both voiding and storage symptoms in patients with BPH {{Chapple, Christopher R

2011}}. Silodosin is a white to pale yellowish-white powder. The partition coefficient [LogP

(octanol/water)] of silodosin is 2.87, with dissociation constants pKa1 of 8.53 and pKa2 of

4.03. According to the US Food and Drug Administration (FDA) label, silodosin is very

slightly soluble in water. The oral bioavailability of silodosin administered as an oral capsule

is nearly 32% and the onset of action (maximum urine flow rate) after the first dose occurs in

2-6 hours {{Montorsi 2010}}. Silodosin undergoes extensive metabolism involving

glucuronidation in the liver {{Matsubara,Y. 2006; Rossi,M. 2010}}.

Because of these issues, an oral film formulation, intended for sublingual administration,

could be promising for silodosin. This would facilitate rapid absorption, provide a faster onset

5

of action, and have potential for faster relief of symptoms than an oral capsule. In addition,

patient compliance could be improved, as some patients find films easier to take than

capsules. Also, the sublingual film formulation of silodosin avoids first-pass metabolism in

the liver and thus has potential to improve systemic bioavailability. Films can also provide a

supersaturated concentration of the drug in the saliva, which can improve the oral mucosal

absorption of poorly soluble drugs. To our knowledge this is the first study of the preparation

of a sublingual film dosage form for silodosin.

The main aim of this study was to prepare oral film formulations of silodosin intended for

sublingual administration. The effects of added polymer and surfactant on the physico-

mechanical and dissolution properties of the films were investigated. The dissolution and

supersaturation properties of the films were investigated in small volumes of simulated saliva

to realistically mimic the oral cavity. HPMC and HPMC-AS were investigated as film-

forming polymer excipients in drug:polymer ratios of 1:3 w/w and 1:5 w/w and tocopherol

polyethylene glycol 1000 succinate (vitamin E; TPGS) was included as a surfactant. The

silodosin dose in each film was 8 mg (the daily recommended dose of silodosin according to

the FDA) and the films were 6 cm2 (2 cm × 3 cm) in area.

2. Materials and methods

2.1. Materials

Silodosin was obtained from Ultra Medica (Damascus, Syria). Hydroxypropyl

methylcellulose 6cp (Pharmacoat 606) and Hydroxypropyl methylcellulose acetate succinate

3cp (Hypromellose Acetate Succinate NF) grade AS-HF were obtained from Shin-Etsu

(Tokyo, Japan). D-α-tocopherol polyethylene glycol 1000 succinate (vitamin E; TPGS) NF

grade was received as a gift sample from BASF Chemicals (Ludwigshafen, Germany).

Glycerol and acesulfame potassium were purchased from VWR chemicals (Stockholm,

Sweden). The water used in all experiments was ultrapure, freshly collected from a Millipore

water system (Milli Q, Sweden). The other materials were purchased locally and used as

purchased.

6

2.2. Methods

2.2.1. Drug-polymer miscibility

Silodosin and the polymers (HPMC or HPMC-AS) were mixed in three ratios (1:1, 1:3 and

1:5 w/w) to a total weight of 400 mg. The solid dispersions were prepared by the film-casting

method. The drug and polymers were dissolved in ethanol:water (1:1 v/v) to a volume of 5

mL. This solution was cast onto a fluoropolymer-coated polyester sheet (Scotchpak® release

liner 1022, 3M Inc., USA) and dried at 70˚C in an oven for 1 hour. The dried samples were

then analyzed using differential scanning calorimetry (DSC; TA Instruments Q 1000, USA) to

investigate the miscibility of the drug and the polymer.

2.2.2. Preparation of the casting gel

The casting gel consisted of HPMC or HPMC-AS (65%), glycerol and propylene glycol (7%),

and sweetener (2%), with ethanol and water as vehicle, relative to the total weight of the solid

base. All weights are w/w ratios. Silodosin was dissolved in ethanol (12-13%) and the

remaining excipients were dissolved in water (87-88%). HPMC or HPMC-AS was gradually

added to this solution under constant magnetic stirring (800 rpm) at ambient temperature (21

± 1 ˚C) until a homogeneous gel was obtained. This casting gel was kept for 6–12 h to remove

the air bubbles. Table 1 shows the overall composition of the prepared films.

2.2.3. Preparation of drug-loaded films

The casting gel (10g) was cast onto a fluoropolymer-coated polyester sheet (Scotchpak®

release liner 1022, 3 MInc., USA) using an automated film applicator equipped with a coating

knife (Coatmaster 510, Erichsen, Sweden). The silodosin dose of 8 mg was loaded into each 6

cm2 film by fixing the wet film thickness at 750μm with a casting speed of 5 mm/s, estimated

from the formula developed by Preis et al, {{ Preis 2012}}. The cast films were dried in a

convective hot-air oven (Binder, Sweden) at 60 °C for 45-50 min. After drying, the films were

carefully peeled off, sealed in plastic (polythene) zip pouches, and stored in a desiccator (23

°C/40% RH) until further characterization.

7

2.2.4. Dry film thickness

The thickness of the films was measured using a Vernier caliper (Cokraft®, Digital caliper,

Sweden). The thickness of each film was measured at the four sides and at the middle point.

The average and standard deviation were calculated.

2.2.5. Differential scanning calorimetry (DSC)

Thermograms of the drug, the drug/polymer blends and the film samples were recorded using

a differential scanning calorimeter (TA Instruments Q 1000, USA) equipped with a

refrigerated cooling system. Each sample (1–3 mg) was placed in a standard aluminum pan

and sealed. The samples were heated at a rate of 10 °C/min from 25 to 120 °C under nitrogen

purge (50 mL/min). The calorimeter was previously calibrated for temperature and heat

capacities using indium and sapphire. The results were analyzed using Universal analysis

software (TA instruments, USA).

2.2.6. Thermo-gravimetric analysis (TGA)

A TGA instrument (TA instruments, USA) was used for thermo-gravimetric analysis and to

determine the moisture content of the films. Approximately 5–8 mg of film (small pieces)

were placed in a platinum pan and heated from 25 to 150 °C at a constant heating rate

(10 °C/min) under nitrogen flow (50 mL/min). The results were analyzed using Universal

analysis software (TA instruments, USA).

2.2.7. Powder X-ray diffraction (PXRD)

PXRD patterns from pure crystalline silodosin and the film samples were collected using an

Empyrean PXRD instrument (PANalytical, Almelo, The Netherlands) equipped with a

PIXel3D detector and monochromatic Cu Kα X-Ray radiation (λ = 1.54056 Å). The voltage

and current were 45 kV and 40 mA. The samples (3 × 3 cm2 films) were placed on a silicone

(zero background) plate which was fitted into the metal sample holder. The samples were

scanned (diffraction angle 2θ) between 5° and 40°, increasing at a step size of 0.02. All

patterns were obtained at 25 ± 1 °C. The data were processed using HighScore Plus software

(PANalytical, The Netherlands).

8

2.2.8. Mechanical properties

The dynamic mechanical strength was tested using a hybrid rheometer in DMA mode (DHR2,

TA Instruments, Sweden). Briefly, samples of cast films were cut into rectangular strips of

1×5 cm2 and 1 cm at each end was held between clamps; thus, the effective testing area was

1×3 cm2. The upper clamp was then used to stretch the film upwards at a constant linear rate

of 0.1 mm/min until the film ruptured. Stress and strain were computed by Trios® software.

The tensile strength (TS) and the elongation at break (EB) were obtained from the peak stress

and the maximum strain, respectively, in the stress vs strain plot. Tensile tests are commonly

used to determine the robustness of film preparations. The TS is the maximum force applied

to the film sample at the breaking point and the EB is the length of the film during the pulling

process. In addition, Young´s modulus or the elastic modulus (EM) describes the influence of

the strain and its force at this strain on the film area. The EM was obtained from the initial

elastic deformation region in the stress vs strain plot {{Radebaugh, Galen W 1988 }}.

𝑇𝑒𝑛𝑠𝑖𝑙𝑒 𝑠𝑡𝑟𝑒𝑛𝑔𝑡ℎ (𝑇𝑆) = Peak stressCross−sectional area of the film

(1)

𝐸𝑙𝑜𝑛𝑔𝑎𝑡𝑖𝑜𝑛 𝑎𝑡 𝑏𝑟𝑒𝑎𝑘 (𝐸𝐵) = Increase in length at breakInitial film length

× 100 (2)

2.2.9. Scanning electron microscopy (SEM)

A Merlin scanning electron microscope (Zeiss, Oberkochen, Germany) equipped with X-Max

50 mm2 X-ray detectors (Oxford Instruments, Abingdon, UK) was used to examine the

morphology of the films. The instrument voltage was 20 kV and the current was 1 nA. The

selected film samples were coated with tungsten before the examination to increase the

conductivity of the electron beam.

2.2.10. HPLC analytical method

The drug content of the films was analyzed in a high performance liquid chromatography

(HPLC) system (Agilent systems Inc., USA) with an auto sampler. The sample separation was

performed on an Agilent Eclipse-plus C18 column (5 μm, 250 mm × 4.6 mm) with a mobile

phase of 25 mM potassium-dihydrogen phosphate buffer (pH 7.0) and acetonitrile 40:60 (v/v)

at 25˚C. The flow rate was 1 mL/min and the determination wavelength was 269 nm {{Runja

2012}}.

9

2.2.11. Drug content

The films (1 × 1 cm2) were placed in a volumetric flask containing 10 mL of water and

ethanol (1:1 v/v) and kept under magnetic stirring at 100 rpm for 1 h. The obtained solution

was filtered through a syringe filter (0.2 μm) and the filtrate was analyzed for drug content

using HPLC.

2.2.12. Disintegration time

Samples (1 × 1 cm2) were placed in a Petri dish containing 2 mL of water and shaken at

60 rpm using an orbital shaker water bath at 37 ± 1 °C. The disintegration time of the films

was evaluated using a modified Petri dish method {{Garsuch, Verena 2010}}. The time to

disintegration or disruption was measured with a stopwatch.

2.2.13. Solubility studies

The solubility of silodosin was determined in simulated saliva containing pre-dissolved

HPMC or HPMC-AS with or without TPGS in concentrations similar to those used in the film

formulations. The simulated saliva was prepared using compositions mentioned by Hobbs and

David {{Hobbs, David 2013}}. An excess of drug was added to conical flasks containing

10 mL of saliva and the other polymer-surfactant excipients. The flasks were tightly closed

and placed in a shaker water bath at 37°C. After 48 h, the separated aliquots were filtered

through a 0.45 μm filter, diluted appropriately and analyzed using HPLC. Each experiment

was performed in triplicate (n = 3) and the results were reported as means ± standard

deviation.

2.2.14. In vitro dissolution in simulated saliva

Non-sink dissolution studies were carried out in simulated saliva at pH 6.8. The films (F1-F8;

2 × 3 cm) were carefully dropped into 10 mL dissolution medium under continuous orbital

shaking (60 rpm) at 37˚C. Samples of the medium were then withdrawn at different times,

filtered through syringe filters (0.45 μm) and analyzed by HPLC.

The degree of supersaturation (DS) was calculated from the drug concentrations at different

times during dissolution of the films (F1-F8) and the drug concentrations at equilibrium. DS

calculations for the formulated drug were based on equation 3:

DSt = CtCeq

(3)

10

Where: Ct is the drug concentration at time t and Ceq is the equilibrium solubility of the drug

in the test medium. The obtained values of DS for F1-F8 were plotted versus time.

2.2.15. Statistical analysis

One-way ANOVA and Tukey's post hoc multiple comparisons were used to determine

statistically significant differences (p < 0.05). All results were expressed as averages plus

standard deviation (n=3). Mechanical properties were investigated using n = 5.

3. Results and Discussion

3.1. Thermal and solid-state properties

DSC thermograms of pure and amorphous silodosin showed a sharp endothermic event at

~106 ˚C, confirming its crystalline state. This is in line with results from a previous report by

Singh and Aniruddh {{Singh, Aniruddh 2016 }}. To choose the best drug:polymer ratio in the

miscible system for forming films, different polymer ratios were investigated (Fig. 1). The

melting peak of the crystalline silodosin (~106 ˚C) was absent in the thermograms of all the

silodosin:polymer systems under investigation, indicating that all the systems were miscible

and formed solid dispersions (Fig. 1). These results confirmed the formation of miscible

dispersions with the studied ratios. Thus, the silodosin:polymer ratios 1:3 and 1:5 were chosen

for film formulations with either HPMC or HPMC-AS, with or without the surfactant (TPGS)

(Table 1). The polymer content was necessary for casting the films and preventing drug

recrystallization during storage.

The thermal properties of pure silodosin and the formulated films were assessed as shown in

Fig. 2. The melting peak of the crystalline form was absent from the DSC thermograms of all

the prepared films (F1-F8), which was interpreted as molecular dispersion of the drug in the

polymer (Fig. 2). ElMeshad and El Hagrasy have also reported the formation of a uniform

dispersion with complete molecular miscibility of different film components in films prepared

with HPMC {{ElMeshad and El Hagrasy, 2011}}. A similar finding of solid dispersions of

amorphous nifedipine in HPMC-AS was reported by Curatolo et al. {{Curatolo 2009}}. A

glass transition temperature (Tg) of 92.9 ˚C for HPMC-based films and 79.28 ˚C for HPMC-

AS-based films was observed (data not shown). The Tg values for the films were higher than

the temperature in the buccal cavity and also the environmental temperature, which is

important for keeping the product stable during storage (from the perspective of the product

11

logistics from manufacturing to consumption){{Vila, Marta MDC 2014; Garsuch, Verena

2010}}.

TGA was performed to determine the moisture content in the prepared films, as shown in Fig.

3. Weight loss was between 0.9 and 1.6 % for the films prepared with the HPMC polymer

(F1-F4) and between 1.6 and 1.9 % for the films prepared with the HPMC-AS polymer. These

results suggest that the film formulations retained some moisture, possibly as a result of the

inherent water sorption properties of both polymers, as suggested by the moisture content in

pure HPMC and HPMC-AS (data not shown). A moisture content of about 2% is essential for

flexibility of the films and this was found not to affect the physical stability of the solid

dispersions, as confirmed by DSC and PXRD analysis (Fig.s 2 and 4). However, it was

observed that the films prepared with HPMC-AS, but not the HPMC films, were tacky.

The solid state of pure silodosin and the film formulations were analyzed using PXRD (Fig.

4). The PXRD pattern of pure silodosin showed sharp, characteristic peaks at 2θ angles of

approximately 10, 11 and 20, illustrating the crystalline nature of the starting material. This is

in agreement with the PXRD patterns reported by Singh and Aniruddh {{Singh, Aniruddh

2016}}. These characteristic peaks disappeared and a hollow shape was observed in the

PXRD patterns for all the HPMC-AS film formulations (F6-F8), which confirms the

formation of a solid dispersion with the drug uniformly dispersed in the polymer. However, in

the case of HPMC films (F1-F4), very low intensity diffraction peaks were observed, which

suggests that the drug may not have been fully dispersed or may have existed as submicron

particles in the polymer matrix.

3.2. Film Morphology

As shown in Fig. 5, the surface morphology of the pure silodosin and representative film

formulations were characterized using SEM. The SEM micrographs of the pure silodosin

show particles with irregular morphology. The HPMC-based films (F2 and F4), but not the

HPMC-AS-based films (F6 and F8), had submicron silodosin particles and tiny pores (Fig 5).

This observation was in agreement with the PXRD results. The submicron particles of

silodosin may have formed at the point of supersaturation in HPMC during preparation of the

casting gel. Alonzo et al. reported the formation of submicron particles in HPMC-based

amorphous solid dispersions that were related to the degree of supersaturation {{Alonzo,

David E 2011}}. This also suggests that more uniform dispersion was obtained in HPMC-AS

12

as a result of the higher solubility of silodosin in HPMC-AS than in HPMC {{Tanno, Fumie

2004; Korhonen, Kristiina 2016;}}.

3.3. Mechanical properties

The mechanical and tensile properties of the thin films were measured under ambient

conditions. The EM, TS and EB of the silodosin films are shown in Table 2. TS was greater in

HPMC-based films than in HPMC-AS-based films, while the EB was longer in HPMC-AS-

based films. Decreasing the content of the polymers reduced the TS of the films. TS and EM

values decreased by 53% and 54%, respectively, in HPMC samples with a drug:polymer ratio

of 1:3 compared with a ratio of 1:5. The effect of less polymer was even more profound with

HPMC-AS; TS and EM decreased by 73% and 86%, respectively. The EB was also affected

by decreasing the proportion of polymer, increasing in HPMC- and HPMC-AS-based films

with a drug:polymer ratio of 1:3 by 24% and 67%, respectively, compared with a ratio of 1:5.

Interestingly, the addition of TPGS had no significant effect on the EM or TS, whereas a

mixed result was seen for the EB. When TPGS was added to the formulations containing the

higher proportion of polymer (drug:polymer ratio 1:5), the EB was increased for HMPC films

and decreased for HMPC-AS films; however, there was no change in EB when TPGS was

added to the films containing the lower proportion of polymer (ratio 1:3). Further studies are

needed to determine the cause of this difference.

The ability to sustain TS is important for packaging and handling the thin films. The obtained

TS values for our HPMC films are similar to those in the literature {{Visser 2015}}. Our TS

values are also comparable to those of the commercial products examined by Pries et al.

{{Preis, Maren 2014}} with respect to maximum force, displacement and elongation,

particularly in comparison with PediaLax and Triaminic® Cold & Cough products. Thus, it

can be inferred that our formulated batches would be suitable for commercialization. In fact,

the elongation properties of the HPMC-AS films considerably exceeded those of the

commercial products; hence films made to this formulation would possess superior toughness.

Tensile properties are a function of the molecular structure. HPMC-AS has fewer polar

substituents, which are known to improve elongation, but also to decrease TS {{Brady, J.

2017}}). In contrast, while TS was higher with HPMC, the films were not as tough, i.e. they

were hard and brittle {{Huang, Yuan 2005}}.

13

3.4. Film thickness, drug content uniformity

The average thickness of the HPMC-based films (F1-F4) ranged from 106.7±0.0 to 116.7±0.0

mm while that of the HPMC-AS-based films (F6-F8) ranged from 86.7±0.0 to 106.7±0.0 mm.

Despite the constant monitoring of processing parameters for all film formulations, there were

differences in the thicknesses of the two types of film. These differences were attributed to

variations in the density of the casting gels for the two polymers as a result of their intrinsic

polymer properties. Differences in film thickness between films with the same polymer-based

formulations may have resulted from the different solid contents and drug:polymer ratios

among the formulations. Evaluation of the drug content in the films showed values ranging

from 96 ± 28 to 117±28.9 %, as shown in Table 1. These results indicated uniform

distribution of drug in the films, within the acceptable limits for standard oral solid dosage

forms, according to the USP {{Liew 2012; Nagy 2010}} .

3.5. Disintegration time:

The disintegration times are shown in Table 2. The fastest disintegration was observed with

the HPMC-AS-based film F5 (drug:polymer ratio 1:5) which disintegrated in 15.3±1.2 sec.

Disintegration was slower for the film based on the HPMC polymer with the same

drug:polymer ratio (35.3±0.6 sec). This effect may have been related to the properties of the

polymer, i.e. wettability and surface tension. Thus, HPMC-AS films disintegrated more

quickly than HPMC films {{Miller-Chou, Beth A 2003; Tanno, Fumie 2004}}. Addition of

the surfactant (TPGS) to the film formulation had an additive effect on the disintegration time

(Table 2). These results were in agreement to those of Vuddanda et al, who found an additive

effect of the surfactant TPGS on speed of disintegration {{Vuddanda 2017}}.

3.6. In vitro dissolution in simulated saliva

The solubility of pure silodosin in simulated saliva was 0.46 mg/mL. The solubility of

silodosin in simulated saliva with additional dissolved HPMC (with and without TPGS)

ranged from 0.48 to 0.51 mg/mL at equilibrium after 48 hours, and with additional dissolved

HPMC-AS (with and without TPGS) ranged from 0.94 to 1.1 mg/mL. The dissolution results

for the films formulated using HPMC and HPMC-AS are shown in Fig.s 6a and 6b,

respectively. In the HPMC-based films, 80% of the drug was released in 10 minutes from F2

(with a drug:polymer ratio of 1:3), while dissolution was poorer for F1 (1:5 drug:polymer

ratio), with 80% release after 25 minutes. This may be because the increase in polymer

14

concentration led to the formation of a gel-like state which decreased water uptake and

retarded drug release, as reported by Singh and Harmanpreet { Singh, Harmanpreet 2017;}}.

Alhayali et al. have previously reported that, in some cases of solid drug dispersions , the drug

concentrations do not change with different drug:polymer ratios {{Alhayali, Amani 2016}}.

Addition of TPGS also improved the drug release noticeably. Thus, films containing TPGS

dissolved faster than TPGS-free films. Other workers have mentioned this effect of the

surfactant TPGS in terms of its ability to improve the solubility, dissolution rate and

bioavailability of some drugs formulated as oral films {{Guo 2013; Vuddanda 2017}}.

In HPMC-AS-based films, about 80% of the drug was released in around 10 minutes (F5 and

F6), with no significant differences in dissolution rate between the two ratios (1:3 and 1:5).

Addition of the surfactant TPGS to the HPMC-AS-based films negatively affected the

dissolution rate for both ratios. Therefore, this study has demonstrated that HPMC works well

as a film-forming polymer when combined with TPGS. In a case study by Garsuch and

Verena {{Garsuch, Verena 2010 }}, HPMC was the most suitable film-forming material of

the excipients tested, providing faster dissolution and easier-to-handle films. Interestingly, the

submicron particles appear not to have affected the faster drug release from HPMC films.

Further, it was observed that the dissolution behavior of HPMC-AS (TPGS-free) films was

similar to that of HPMC films containing amorphous solid dispersions (Fig. 6).

The dissolution studies were conducted in simulated saliva (pH 6.8) under non-sink

conditions, mimicking the conditions of the oral cavity. These studies suggested that both

polymers are capable of increasing and prolonging the supersaturation of the drug and helping

to prevent the tendency of the drug to precipitate and recrystallize during dissolution. This

could be attributed to the existence of the drug in the amorphous state and molecularly

dispersed in the polymer matrix. This was also evident from the thermal, solid-state and

morphological results.

It has been reported that cellulose-derived polymers such as HPMC, HPMC-AS and

hypromellose phthalate (HPMCP) are superior for preparing solid, amorphous dispersions,

particularly with poorly water-soluble drugs, compared to other polymers. The bulky structure

of the polymer network and worse solubility properties (compared to highly water-soluble

polymers) facilitate the reduction of drug mobility in the polymer matrix and prevent the drug

from recrystallizing as normally induced by supersaturation during dissolution (Li and Taylor,

2018). This consequently improves the product’s physical stability and in vitro drug release

15

performance. In general, our dissolution results revealed that HPMC and HPMC-AS would be

useful for preparing films intended for oral cavity absorption, which is more complex than

GIT absorption.

The DS values for silodosin formulated as an oral film are presented in Fig. 7. The highest DS

was obtained for HPMC-based formulations at early time points. F4, in particular, had a DS

value of > 1 (supersaturated) before 5 minutes, which was earlier than the other films

formulated using the HPMC polymer (F1-F3), as shown in Fig. 7. In the dissolution results,

80% of the drug was released from film formulation F4 during the first 10 minutes (Fig. 6).

This effect could be attributed to the synergistic effect of the surfactant (TPGS) and the

polymer (HPMC), resulting in improved solubility and maintained supersaturation {{Kim,

M.S.2014; Prasad, Dev 2014; Gao, Ping 2004}}. Therefore, film formulation F4 appears to

have potential for the development of a silodosin film formulation with improved

performance and improved oral sublingual absorption as a result of the high degree of

supersaturation solubility {{Brouwers, J. 2009; Yang, Meiyan 2016}}.

4. Conclusions

Oral films of silodosin intended for the sublingual administration route were prepared

successfully for the first time. DSC studies confirmed the existence of silodosin in an

amorphous form in films formulated with HPMC or HPMC-AS. SEM and PXRD studies

revealed the presence of submicron particles of the drug in HPMC-based films, while the drug

remained fully amorphous in HPMC-AS films. The mechanical properties of HPMC films

were better than those of HPMC-AS films with respect to stability during patient handling and

packing. The dissolution behavior of HPMC was similar to that of HPMC-AS when the

surfactant TPGS was added (0.5 % w/w) to the HPMC film formulation. TPGS at the tested

concentration had no effect on the dissolution of the drug in HPMC-AS-based formulations.

Silodosin formulation F4 had a DS >1, which could be promising for improving its oral

absorption. Further studies are required to evaluate the dissolution of the film in human saliva,

and to investigate the permeability of the oral cavity to the drug and the drug absorption

characteristics. Film palatability and crystallization during storage (stability) also require

investigation.

16

Acknowledgments

A. Alhayali is grateful for the financial support from the Iraqi Ministry of Higher Education

and Scientific Research (MOHESR) for PhD grant. M. Elbadawi acknowledged for his

assistance and advice about mechanical properties section.

17

Table1. Silodosin oral films. Summary of the drug:polymer ratios for the eight developed

films, and the mean thickness, speed of disintegration and drug content

Formulation

code

Drug:polymer

ratio

Thickness

(mm)

Disintegration time

(seconds)

Drug

content

(mg)

F1 1:5 (Silodosin:HPMC) 106.7±0.0 35.3±0.6 96±28

F2 1:3 (Silodosin:HPMC) 110±0.0 61.0±0.0 97 ±5.8

F3* 1:5 (Silodosin:HPMC) 116.7±0.0 65.7±0.6 96±1.1

F4* 1:3 (Silodosin:HPMC) 113±0.0 62.7±1.5 104 ±26.1

F5 1:5 (Silodosin:HPMC-AS) 106.7±0.0 15.3±1.2 98 ±6.5

F6 1:3 (Silodosin:HPMC-AS) 86.7±0.0 33.7±2.5 113±18.4

F7* 1:5 (Silodosin:HPMC-AS) 90±0.0 33.0±1.0 117±28.9

F8* 1:3 (Silodosin:HPMC-AS) 86.7±0.0 56.7±0.6 99±19.8

Mean ± Standard deviation (n ≥ 3).

HPMC = hydroxypropyl methylcellulose; HPMC-AS = hydroxypropyl methylcellulose

acetate succinate.

*Four films (F3, F4, F7 and F8) also contained the surfactant tocopherol polyethylene glycol

succinate.

18

Table 2: Mechanical properties of silodosin films

Film

(F)

Young´s Modulus

(Mpa)

Max. Tensile

Strength

(Mpa)

Elongation at Break

(%)

1 490±63 10.08±1.59 3.82±0.96

2 225±120 4.74±2.66 4.74±2.21

3 458±34 9.81±0.67 6.35±0.45

4 158±17 3.29±0.25 4.44±0.71

5 337±25 7.31±3.01 12.31±1.33

6 48±19 1.94±0.29 20.57±3.95

7 314±25 6.91±0.96 8.53±1.16

8 54±11 1.99±0.17 20.98±4.33

Means ± Standard deviation (n ≥ 5)

19

Figure 1: Differential scanning calorimetry results for drug-polymer miscibility

determination.

20

Figure 2: Differential scanning calorimetry results for the silodosin film formulations. (a)

HPMC-based films, F1-F4; and (b) HPMC-AS-based films, F5-F8.

21

Figure 3: Thermogravimetric analysis results for the formulated silodosin films showing

moisture content.

22

Figure 4: X-ray diffraction patterns for pure crystalline silodosin and the developed films

(F1-F8).

23

Figure 5: Scanning electron microscopy micrographs for the formulated films and the pure

drug (silodosin). The bar represents 100 μm for F2, F4, F6, and F8, and 20 μm for pure

silodosin.

24

Figure 6: Film dissolution in simulated saliva for (a) the HPMC-based films and (b) the

HPMC-AS-based films. n=3 ± Standard deviation

25

Figure 7. Degree of supersaturation as a result of formulating the drug silodosin in film

formulations containing HPMC (F1-F4) or HPMC-AS (F5-F8). n= 3± Standard deviation

26

References

1. Alhayali, A.; Tavellin, S.; Velaga, S. Dissolution and precipitation behavior of ternary solid

dispersions of ezetimibe in biorelevant media. Drug Dev. Ind. Pharm. 2016, 1-10.

2. Allam, A.; Fetih, G. Sublingual fast dissolving niosomal films for enhanced bioavailability

and prolonged effect of metoprolol tartrate. Drug Des. Devel. Ther. 2016, 10, 2421-2433.

3. Alonzo, D. E.; Gao, Y.; Zhou, D.; Mo, H.; Zhang, G. G.; Taylor, L. S. Dissolution and

precipitation behavior of amorphous solid dispersions. J. Pharm. Sci. 2011, 100, 3316-3331.

4. Boateng, J. S.; Auffret, A. D.; Matthews, K. H.; Humphrey, M. J.; Stevens, H. N.;

Eccleston, G. M. Characterisation of freeze-dried wafers and solvent evaporated films as

potential drug delivery systems to mucosal surfaces. Int. J. Pharm. 2010, 389, 24-31.

5. Brady, J.; Dürig, T.; Lee, P. I.; Li, J. -. In Chapter 7 - Polymer Properties and

Characterization; Qiu, Y., Chen, Y., Zhang, G. G. Z., Yu, L. and Mantri, R. V., Eds.;

Developing Solid Oral Dosage Forms (Second Edition); Academic Press: Boston, 2017; pp

181-223.

6. 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.

7. Campisi, G.; Giannola, L. I.; Florena, A. M.; De Caro, V.; Schumacher, A.; Göttsche, T.;

Paderni, C.; Wolff, A. Bioavailability in vivo of naltrexone following transbuccal

administration by an electronically-controlled intraoral device: a trial on pigs. J. Controlled

Release 2010, 145, 214-220.

8. Chapple, C. R.; Montorsi, F.; Tammela, T. L.; Wirth, M.; Koldewijn, E.; Fernández, E. F.;

European Silodosin Study Group Silodosin therapy for lower urinary tract symptoms in men

with suspected benign prostatic hyperplasia: results of an international, randomized, double-

blind, placebo-and active-controlled clinical trial performed in Europe. Eur. Urol. 2011, 59,

342-352.

9. Curatolo, W.; Nightingale, J. A.; Herbig, S. M. Utility of hydroxypropylmethylcellulose

acetate succinate (HPMCAS) for initiation and maintenance of drug supersaturation in the GI

milieu. Pharm. Res. 2009, 26, 1419-1431.

27

10. De Caro, V.; Giandalia, G.; Siragusa, M.; Sutera, F.; Giannola, L. New prospective in

treatment of Parkinson's disease: Studies on permeation of ropinirole through buccal

mucosa. Int. J. Pharm. 2012, 429, 78-83.

11. ElMeshad, A. N.; El Hagrasy, A. S. Characterization and optimization of orodispersible

mosapride film formulations. Aaps Pharmscitech 2011, 12, 1384-1392.

12. El-Setouhy, D. A.; El-Malak, N. S. A. Formulation of a novel tianeptine sodium

orodispersible film. Aaps Pharmscitech 2010, 11, 1018-1025.

13. Gao, P.; Guyton, M. E.; Huang, T.; Bauer, J. M.; Stefanski, K. J.; Lu, Q. Enhanced oral

bioavailability of a poorly water soluble drug PNU‐91325 by supersaturatable

formulations. Drug Dev. Ind. Pharm. 2004, 30, 221-229.

14. Garsuch, V.; Breitkreutz, J. Comparative investigations on different polymers for the

preparation of fast‐dissolving oral films. J. Pharm. Pharmacol. 2010, 62, 539-545.

15. Guo, Y.; Luo, J.; Tan, S.; Otieno, B. O.; Zhang, Z. The applications of Vitamin E TPGS in

drug delivery. European Journal of Pharmaceutical Sciences 2013, 49, 175-186.

16. Hanif, M.; Zaman, M.; Chaurasiya, V. Polymers used in buccal film: a review. Designed

Monomers and Polymers 2015, 18, 105-111.

17. Hobbs, D.; Karagianis, J.; Treuer, T.; Raskin, J. An in vitro analysis of disintegration

times of different formulations of olanzapine orodispersible tablet: A preliminary

report. Drugs R. 2013, 13, 281-288.

18. Huang, Y.; Liu, J.; Zhang, Z. Synthesis and characterization of HPMC derivatives as

novel duodenum-specific coating agents. Arch. Pharm. Res. 2005, 28, 364-369.

19. Kalia, V.; Garg, T.; Rath, G.; Goyal, A. K. Development and evaluation of a sublingual

film of the antiemetic granisetron hydrochloride. Artificial cells, nanomedicine, and

biotechnology 2016, 44, 842-846.

20. Kapoor, A. Benign prostatic hyperplasia (BPH) management in the primary care

setting. Can. J. Urol. 2012, 19, 10-17.

28

21. Kim, M. S.; Baek, I. H. Fabrication and evaluation of valsartan-polymer- surfactant

composite nanoparticles by using the supercritical antisolvent process. Int. J.

Nanomedicine 2014, 9, 5167-5176.

22. Koland, M.; Sandeep, V.; Charyulu, N. Fast dissolving sublingual films of ondansetron

hydrochloride: effect of additives on in vitro drug release and mucosal permeation. Journal of

Young Pharmacists 2010, 2, 216-222.

23. Korhonen, K.; Poikolainen, M.; Korhonen, O.; Ketolainen, J.; Laitinen, R. Systematic

evaluation of a spraying method for preparing thin Eudragit-drug films by Design of

Experiments. Journal of Drug Delivery Science and Technology 2016, 35, 241-251.

24. Li, N.; Taylor, L. S. Tailoring supersaturation from amorphous solid dispersions. J.

Controlled Release 2018.

25. Liew, K. B.; Tan, Y. T. F.; Peh, K. K. Characterization of oral disintegrating film

containing donepezil for Alzheimer disease. AAPS PharmSciTech 2012, 13, 134-142.

26. Madhav, N. S.; Shakya, A. K.; Shakya, P.; Singh, K. Orotransmucosal drug delivery

systems: a review. J. Controlled Release 2009, 140, 2-11.

27. Mashru, R.; Sutariya, V.; Sankalia, M.; Parikh, P. Development and evaluation of fast-

dissolving film of salbutamol sulphate. Drug Dev. Ind. Pharm. 2005, 31, 25-34.

28. Matsubara, Y.; Kanazawa, T.; Kojima, Y.; Abe, Y.; Kobayashi, K.; Kanbe, H.; Harada,

H.; Momose, Y.; Terakado, S.; Adachi, Y.; Midgley, I. Pharmacokinetics and disposition of

silodosin (KMD-3213). Yakugaku Zasshi 2006, 126 Spec no., 237-245.

29. Miller-Chou, B. A.; Koenig, J. L. A review of polymer dissolution. Progress in Polymer

Science 2003, 28, 1223-1270.

30. Montorsi, F. Profile of silodosin. European Urology Supplements 2010, 9, 491-495.

31. Nagy, Z. K.; Nyúl, K.; Wagner, I.; Molnár, K.; Marosi, G. Electrospun water soluble

polymer mat for ultrafast release of Donepezil HCl. Express polymer letters 2010, 4, 763-772.

32. Patel, V. F.; Liu, F.; Brown, M. B. Advances in oral transmucosal drug delivery. J.

Controlled Release 2011, 153, 106-116.

29

33. Prasad, D.; Chauhan, H.; Atef, E. Amorphous stabilization and dissolution enhancement

of amorphous ternary solid dispersions: combination of polymers showing drug–polymer

interaction for synergistic effects. J. Pharm. Sci. 2014, 103, 3511-3523.

34. Preis, M.; Knop, K.; Breitkreutz, J. Mechanical strength test for orodispersible and buccal

films. Int. J. Pharm. 2014, 461, 22-29.

35. Preis, M.; Pein, M.; Breitkreutz, J. Development of a taste-masked orodispersible film

containing dimenhydrinate. Pharmaceutics 2012, 4, 551-562.

36. Preis, M.; Woertz, C.; Kleinebudde, P.; Breitkreutz, J. Oromucosal film preparations:

classification and characterization methods. Expert Opinion on drug delivery 2013, 10, 1303-

1317.

37. Radebaugh, G. W.; Murtha, J. L.; Julian, T. N.; Bondi, J. N. Methods for evaluating the

puncture and shear properties of pharmaceutical polymeric films. Int. J. Pharm. 1988, 45, 39-

46.

38. Repka, M. A.; Prodduturi, S.; Stodghill, S. P. Production and characterization of hot-melt

extruded films containing clotrimazole. Drug Dev. Ind. Pharm. 2003, 29, 757-765.

39. Rossi, M.; Roumeguere, T. Silodosin in the treatment of benign prostatic

hyperplasia. Drug Des. Devel. Ther. 2010, 4, 291-297.

40. Runja, C.; Pigili, R. DEVELOPMENT AND VALIDATION OF RP-HPLC METHOD

FOR ESTIMATION OF SILODOSIN IN BULK AND PHARMACEUTICAL DOSAGE

FORMS. International Journal of Pharmaceutical Sciences Review & Research 2012, 16.

41. Sharma, R.; Singh, H.; Joshi, M.; Sharma, A.; Garg, T.; Goyal, A. K.; Rath, G. Recent

advances in polymeric electrospun nanofibers for drug delivery. Critical Reviews™ in

Therapeutic Drug Carrier Systems 2014, 31.

42. Singh, A.; Mirmehrabi, M. Crystalline forms of silodosin 2016.

43. Singh, B.; Garg, T.; Goyal, A. K.; Rath, G. Recent advancements in the cardiovascular

drug carriers. Artificial cells, nanomedicine, and biotechnology 2016, 44, 216-225.

44. Singh, H.; Narang, J. K.; Singla, Y. P.; Narang, R. S.; Mishra, V. TPGS stabilized

sublingual films of frovatriptan for the management of menstrual migraine: Formulation,

30

design and antioxidant activity. Journal of Drug Delivery Science and Technology 2017, 41,

144-156.

45. Singh, H.; Sharma, R.; Joshi, M.; Garg, T.; Goyal, A. K.; Rath, G. Transmucosal delivery

of Docetaxel by mucoadhesive polymeric nanofibers. Artificial cells, nanomedicine, and

biotechnology 2015, 43, 263-269.

46. Slavkova, M.; Breitkreutz, J. Orodispersible drug formulations for children and

elderly. European Journal of Pharmaceutical Sciences 2015, 75, 2-9.

47. Tanno, F.; Nishiyama, Y.; Kokubo, H.; Obara, S. Evaluation of hypromellose acetate

succinate (HPMCAS) as a carrier in solid dispersions. Drug Dev. Ind. Pharm. 2004, 30, 9-17.

48. Vila, M. M.; Tardelli, E. R.; Chaud, M. V.; Tubino, M.; Balcão, V. M. Development of a

buccal mucoadhesive film for fast dissolution: mathematical rationale, production and

physicochemical characterization. Drug Deliv. 2014, 21, 530-539.

49. Visser, J. C.; Dohmen, W. M.; Hinrichs, W. L.; Breitkreutz, J.; Frijlink, H. W.;

Woerdenbag, H. J. Quality by design approach for optimizing the formulation and physical

properties of extemporaneously prepared orodispersible films. Int. J. Pharm. 2015, 485, 70-

50. Vuddanda, P. R.; Alomari, M.; Dodoo, C. C.; Trenfield, S. J.; Velaga, S.; Basit, A. W.;

Gaisford, S. Personalisation of warfarin therapy using thermal ink-jet printing. European

Journal of Pharmaceutical Sciences 2018, 117, 80-87.

51. Vuddanda, P. R.; Montenegro-Nicolini, M.; Morales, J. O.; Velaga, S. Effect of

surfactants and drug load on physico-mechanical and dissolution properties of nanocrystalline

tadalafil-loaded oral films. European Journal of Pharmaceutical Sciences 2017, 109, 372-

380.

52. Yang, M.; Gong, W.; Wang, Y.; Shan, L.; Li, Y.; Gao, C. Bioavailability Improvement

Strategies for Poorly Water-Soluble Drugs Based on the Supersaturation Mechanism: An

Update. Journal of Pharmacy & Pharmaceutical Sciences 2016, 19, 208-225.

53. Zhang, H.; Zhang, J.; Streisand, J. B. Oral mucosal drug delivery. Clin.

Pharmacokinet. 2002, 41, 661-680.

Amani I. Y. Alhayali was born in Mosul, Iraq, on November 15, 1976, to Suhailah Yousif and Ibraheem Y. Alhayali. She graduated from Mosul University, Iraq with a Master degree in Clinical Pharmacy in 2004. The dissertation work was carried out at the Department of Health Sciences, Luleå University of Technology with Professor Sitaram Velaga.