DRUG DELIVERY: FUNDAMENTALS AND APPLICATIONS

Ideally, drug development would involve the selection of active pharmaceutical ingredients (APIs) that possess ideal drug delivery characteristics, followed by their development using simple dosage forms. However, the reality is that increasingly formulators must work with APIs that have challenging physicochemical properties, including poor water solubility.

The increase in proportion of poorly soluble candidates is attributed to both improvements in synthesis technology, which has enabled the design of very complicated compounds, and also a change in focus in the discovery strategy of new APIs, from a so-called phenotypic approach to a target-based approach. The phenotypic approach involves trial-and-error methodology in which compounds are tested against cells, tissues, or the whole body. This approach takes into account various physicochemical and biological factors that may affect the efficacy of candidates, including solubility, protein binding, and metabolism. In the target-based approach, candidate compounds are screened against specific targets, based on hypotheses concerning action mechanisms. Lead compounds are typically dissolved in dimethyl sulfoxide for high-throughput screening (HTS), which means that even very poorly soluble drugs can be tested. Although the HTS approach provides a clear lead with respect to molecular design, compounds with poor aqueous solubility can progress to development after screening.

Poor water solubility has important ramifications for the drug discovery process, as poorly soluble lead compounds cannot be adequately formulated for subsequent preclinical studies in animals. Thus, it may not be possible to follow up potentially promising leads, which instead have to be dropped from the discovery process, never realizing their true potential. Although it may be possible to overcome the solubility problem by chemical modification of the drug, in many cases this is not feasible.

Poor water solubility also has important ramifications for drug bioavailability. In order to cross an epithelial interface, the drug must usually be dissolved in the biological fluids at that interface. For example, for the oral route, the first step in the oral absorption process is dissolution of the drug in the gastrointestinal (GI) lumen contents. A drug that is poorly soluble in the aqueous GI fluids will demonstrate poor and erratic dissolution, with concomitant low absorption and thus poor bioavailability—even if the drug possesses good intestinal permeability characteristics. Furthermore, the rate of intestinal absorption is driven by the concentration gradient between the intestinal lumen and the blood. A low concentration gradient is a poor driver for absorption, with a concomitant retarded flux across the intestinal epithelium.

As described in detail in Chapter 7, a significant hurdle associated with the oral route is the extreme variability in GI conditions, which can cause large intra- and interindividual variability in pharmacokinetic profiles. Poor water solubility exacerbates this variability, as there is a lack of dose proportionality for these compounds, as well as significant variability depending on the presence of food and fluids in the GI tract. The activity of bile salts on drug solubilization is a further important variable. These formulation and bioavailability concerns are equally relevant for poorly soluble drugs delivered via alternative epithelial routes, such as the pulmonary, topical, nasal, vaginal, and ocular routes.

3.1.1 BIOPHARMACEUTICS CLASSIFICATION SYSTEM

The Biopharmaceutics Classification System (BCS) classifies drugs into four categories, based on their aqueous solubility and ability to permeate the GI membrane (Figure 3.1). (However, it should be noted that the BCS is relevant to permeation across all biological membranes, not just the GI tract.) Based on pioneering work by Gordon Amidon at the University of Michigan (Amidon et al. 1995), the system has been adopted by the U.S. Food and Drug Administration (FDA) to allow pharmaceutical companies a waiver of clinical bioequivalence studies (a biowaiver), when seeking regulation of postapproval changes and generics. Increasingly, the BCS is being used as a tool in product development, to flag up potential solubility and permeability difficulties that may be associated with lead compounds.

A drug is considered highly soluble if its highest dose strength is soluble in less than 250 mL of water, as tested over a pH range of 1–7.5. A drug is considered high permeable if the oral absorption compares favorably (i.e., higher than 90%) to an intravenous injection of the drug. Absorption in vivo can be carried out by monitoring the appearance of the drug in the plasma after oral administration. Intestinal permeability may also be assessed by other methods, including in vivo intestinal perfusions studies in humans, in vivo or in situ intestinal perfusion studies in animals, in vitro permeation experiments with excised human or animal intestinal tissue, and in vitro permeation experiments across epithelial cell monolayers, such as the Caco-2 cell line.

FIGURE 3.1 The Biopharmaceutics Classification System. (Courtesy of Particle Sciences, Inc.)

Using the BCS, four distinct classes of drug can be defined as the following:

• Class I drugs possess characteristics that ensure good bioavailability: they dissolve rapidly in the GI fluids and then rapidly permeate the epithelial barrier.

• Drugs that fall into Class II possess good permeability characteristics, but they have poor solubility, which limits their bioavailability. Approximately 35%–40% of the top 200 drugs listed in the United States and other countries as immediate-release oral formulations are practically insoluble (see also Chapter 20, Figure 20.6). The bioavailability of a Class II drug can be markedly improved by improving its solubility: various methods to improve drug solubility are the focus of this chapter. The fact that about 40% of the top marketed drugs are practically insoluble, yet are nevertheless used commercially, is a testimony to the success of current solubilization methods.

• Class III drugs, although highly soluble, possess poor permeability. Permeability across epithelial barriers and strategies to improve epithelial permeability are described in Chapter 4.

• Class IV drugs have both poor solubility and permeability. In the case of Class IV drugs, improving the solubility may help somewhat toward improving bioavailability, although poor permeability will still be an issue.

3.1.2 STRATEGIES TO IMPROVE WATER SOLUBILITY

Poor solubility and permeability problems may be addressed at the chemical level, via lead optimization: this approach is described in Chapter 20 (Section 20.8). This chapter describes approaches used to increase the solubility of a poorly soluble API. A wide range of approaches can be used, as summarized in Table 3.1.

Which approach to use is partly determined by the nature of the drug. Poorly soluble drugs, i.e., Class II and Class IV of the BCS classification situation, can be further subclassified into two types of molecules (Bergström et al. 2007):

1. “Grease ball”: highly lipophilic compounds, with a high log P (>4) and a low melting point (<200°C). These compounds cannot form bonds with water molecules; thus, their solubility is limited by the solvation process.

2. “Brick dust”: compounds usually with low energy, highly stable crystal forms, with a high melting point (>200°C), and with poor water and lipid solubility (log P < 2). The water solubility of such compounds is restricted due to strong intermolecular bonds within the crystal structure.

TABLE 3.1
Mechanisms to Improve the Solubility of Poorly Soluble Active Pharmaceutical Ingredients

Approach	Example
Physical modification	Micronization
	Nanosizing
Chemical modification	Prodrug formation
Crystal engineering	Salt formation
	Polymorphs
	Cocrystals
Formulation approaches	Solvent composition
	• Cosolvents • pH adjustment • Formulation excipients: surfactants, oils, etc.
	ASDs
	Cyclodextrin inclusion complexes
	Colloidal systems
	• Micelles • SMEDDS • Liposomes • Nano- and microparticulates
ASDs, amorphous solid dispersions; SMEDDS, self-microemulsifying drug delivery systems.

The solubility of “grease ball” molecules can be increased if appropriate formulation strategies are used to overcome, or at least improve, the solvation process. A traditional approach for parenteral formulations is to administer the drug with a cosolvent: as a mixture of water with a water-miscible solvent such as propylene glycol, ethanol, and poly(ethylene glycol) (PEG) 300. However, even with the use of a cosolvent, it may only be possible to achieve low drug loading. Additionally, harsh vehicles such as organic solvents carry the risk of toxicity, particularly cardiotoxicity, in vivo. Precipitation of the drug on dilution with the body fluids may also occur, causing pain and inflammation at the injection site, as well as the possibility of emboli. For intramuscular delivery, precipitation of the formulation may result in the formulation acting more like a depot injection, resulting in the delayed absorption of the drug. Other formulation strategies for “grease ball” molecules include the use of cyclodextrin (CD) inclusion complexes, and the use of micelle and emulsion-based delivery systems. In contrast, “brick dust” molecules are not only poorly soluble in water, but also in oils, rendering them unsuitable for many lipid-based formulation approaches. For “brick dust” molecules, the main strategies used tend to focus on crystal modification, including salt formation and cocrystals.

A further approach is to enhance drug dissolution kinetics. Drug solubility is an equilibrium measure; the rate at which solid drug, or drug in a formulation, passes into solution (i.e., the dissolution rate) is also a critically important parameter. Because intestinal transit time is relatively rapid, a drug with a very slow dissolution rate may not have sufficient time to dissolve in the GI fluids for absorption to take place. Increasing the dissolution kinetics can therefore result in an improved bioavailability for oral formulations. Mechanisms of improving drug dissolution include reducing particle size (e.g., NanoCrystals^®), selecting a metastable polymorphic form, and using amorphous solid dispersions (ASDs).

All these approaches are described in this chapter, beginning with a discussion on particle-size reduction technologies.

3.2 CRYSTAL SIZE: NANOSIZING

According to the Noyes–Whitney equation (Chapter 2, Equation 2.1), the dissolution rate of drug particles is proportional to the surface area of the particles in contact with the dissolution medium. A decrease in drug crystal size results in an increased surface area to volume ratio (Figure 3.2); therefore, reducing the crystal size of a drug powder will increase its dissolution rate. For example, if the crystal size is reduced from 1 μm to 100 nm, the surface area increases 10-fold, which should lead to a 10-fold enhancement of the dissolution rate. For drugs where dissolution is the rate-limiting step, improved dissolution in vivo will translate to higher, and more uniform, bioavailability. Furthermore, as described by the Prandtl equation, the diffusion layer thickness around the drug crystal may also be decreased, thus resulting in an even faster dissolution rate.

FIGURE 3.2 Mechanisms to increase solubility/dissolution rate by nanocrystal formation.

The solubility per se also increases as a result of reducing crystal size. Assuming that a particle is spherical, dependence of solubility on particle size can be described by the Ostwald–Freundlich equation:

$C (r) = C (\infty) \exp (\frac{2 γ M}{r ρ R T})$

(3.1)

where

C(r) and C(∞) are the solubilities of a particle of radius r and of infinite size

γ, M, and ρ are the interfacial tension at the particle surface, the molecular weight of the solute, and the density of the particle, respectively

According to this equation, solubility increases with a decrease in particle size, i.e., an increase in surface curvature. However, an example of the calculation in Figure 3.2, in which typical values are substituted for Equation 3.1, shows that an increase in solubility is almost negligible for crystals of 100 nm (1.02-fold) and only 1.27-fold at 10 nm. A crystal size of 10 nm cannot be produced with current formulation technologies. Additionally, the API crystal may not be spherical, whereas this effect is only valid when an increase in surface curvature is achieved.

Nanocrystals produced by top-down technologies (see Section 3.2.2.1), should also expose high-energy surfaces to the outer environment. This further enhances the dissolution rate (Figure 3.2). This mechanism is not applicable for nanocrystals produced by bottom-up procedures.

An increase in the surface area is thus likely to be the dominant mechanism for the enhanced dissolution in most cases. Micronization, the process of reducing drug crystal to the micron size range via milling techniques, has long been used in the pharmaceutical industry as a means of improving drug formulation and oral bioavailability. Recent improvements in the technology now allows for size reductions to extend even further: “nanosizing” refers to API crystal size reduction down to the nanometer range, typically ca. 100–300 nm, thereby providing a considerable increase in surface area and thus dissolution rate. “Nanomaterials” are defined by the FDA as materials with a length scale of approximately one to one hundred nanometers in any dimension. Although current commercial nanosized preparations typically have a size range of 100–300 nm and are thus outside this range, exceptions are also accepted if the material exhibits dimension-dependent properties or phenomena.

In this chapter, the authors use the term “nanocrystal” in the general sense, to describe any drug crystal in the nanometer size range, whereas NanoCrystals^® denote a patented technology of Elan Corporation, described in Section 3.2.2.1. Once a drug is nanosized, the nanocrystals can be formulated into various dosage forms, including injectables, tablets, capsules, and powders for inhalation; they are thus suitable for delivery via a wide variety of routes. Dispersions of nanocrystals liquid media are known as “nanosuspensions.” Nanosuspensions prepared in water can be used as granulation fluids for the preparation of tablets; produced in oils, they can be used directly to fill capsules. Nanosuspensions may be stored in the liquid form, but postpreparation workup is also possible, such as spray-drying and freeze-drying, to obtain nanocrystals in a dry powder form. Sugars may also be added to formulations, to function as protectants during the drying process.

Nanosized formulations on the market include Avinza^® (morphine sulfate), Focalin^® XR (dexmethylphenidate hydrochloride), Megace ES^® (megestrol acetate), Ritalin^® LA (methylphenidate hydrochloride), Tricor^® (fenofibrate), Triglide^® (fenofibrate), and Zanaflex Capsules^® (tizanidine hydrochloride).

3.2.1 STABILIZATION

Since nanocrystals have high-energy surfaces, stabilizers are needed to prevent irreversible aggregation. Steric stabilization can be achieved by the adsorption of hydrophilic polymers and/or surfactants onto the particle surface. Common polymers and/or surfactants used to provide steric stabilization include poly(vinylpyrrolidone) (PVP), hydroxypropyl methylcellulose (HPMC), hydroxypropylcellulose (HPC), a-tocopheryl PEG-1000-succinate (TPGS), polysorbate (Tween 80), and the pluronic surfactants F68 and F127. The repulsive steric layer prevents the particles from approaching each other. However, in most cases, the use of steric stabilization alone is not sufficient for nanosuspension stability. Further stability can be conferred by the adsorption of charged molecules, such as ionic surfactants (sodium lauryl sulfate and docusate sodium), onto the particle surface. In this case, electrostatic charge repulsion provides an electrostatic potential barrier to particle aggregation. Surfactants also help in the wetting and dispersion of the drug particles, which are usually very hydrophobic. Nanocrystals may have advantages over other surfactant-containing formulations such as emulsions and micelles, because the level of surfactant required is much lower, being only the amount necessary to stabilize the solid–fluid interface.

3.2.2 MANUFACTURE

3.2.2.1 Top-Down Technologies

“Top-down” technologies involve disintegration methods, i.e., starting with coarse crystals and applying forces to reduce them to the nanocrystal range. As stated earlier, these top-down technologies expose high-energy surfaces to the outer environment, which further enhances the dissolution rate (Figure 3.2). The two most widely used technologies are media milling and homogenization.

Media milling. A milling chamber is initially charged with milling media: tiny balls, typically 1 mm or less, comprising materials such as ultradense ceramic media, or glass beads. A suspension of the API, with appropriate stabilizing agents, is then added to the chamber, and the chamber is rotated at a very high shear rate under a controlled temperature. The forces generated from the impaction of milling media with the drug cause crystal disintegration to the nanosize range. Processing time, as well as other operational parameters (milling speed, media load, media size, temperature, additives, etc.) can be tailored in order to maximize the crystal-size reduction process for each particular API. The process can produce stable, nanosized dispersions, with very tight, monodisperse, crystal-size distribution profiles. However, this method carries the risk of media ball erosion, which could result in the presence of unwanted media residues in the final product. For this reason, high-abrasion-resistance balls are used, and the final nanosuspension must be analyzed to ensure the absence of trace impurities.

NanoCrystals^® are a proprietary wet milling technology from the Elan Corporation, which uses a highly crossed-linked polystyrene resin (PollyMill^®) as the milling media (Merisko-Liversidge and Liversidge 2011). A crude slurry of the poorly water-soluble API, in a water-based stabilizer solution, is then added and subjected to shear forces. The NanoCrystal^® particles of the drug are stabilized against agglomeration by the surface adsorption of patented, generally regarded as safe (GRAS), stabilizers.

The NanoCrystal^® technology is used in a variety of commercially available preparations, where reformulation of the poorly soluble drug has resulted in many advantages. Sirolimus is a water-insoluble immunosuppressant drug, which was originally marketed as a self-emulsifying oral solution that required refrigeration and necessitated a complicated reconstitution procedure. In contrast, the NanoCrystal^® tablet formulation (Rapamune^®) offers improved bioavailability, less fluctuations in blood levels, easier storage (as no refrigeration is required), and improved palatability. Another example is the new chemical entity MK-0869, which was developed to treat chemotherapy-induced nausea and vomiting. The NanoCrystal^® formulation (Emend^®) resulted in a 600% improvement in bioavailability. Furthermore, in contrast to the original formulation, there is no need to take the drug with food (an important issue for this patient group, who are suffering from nausea and vomiting). Invega Sustenna^® is the first commercial depot formulation product using NanoCrystal^® technology, for the delivery of paliperidone palmitate in the management of schizophrenia.

High-pressure homogenization. This process involves the application of high shear and impaction forces to drug suspensions in order to reduce their particle size to the nanoscale. A number of commercial technologies now exist, including SkyePharma’s Dissocubes^®, which can produce stable nanoparticle suspensions in water at room temperature. Triglide^® (fenofibrate) is manufactured using this technology. The Nanopure^® technology (PharmaSol GmbH) enables the production of nanosuspensions in nonaqueous media, for example with oils and PEG.

3.2.2.2 Bottom-Up Technologies

Nanocrystals may also be produced via “bottom-up” technologies, which involve assembly methods, i.e., starting from molecules in solution, then building up to form solid nanocrystals. Note that bottom-up procedures may produce metastable crystalline forms, including the amorphous state; the advantages and disadvantages of which are discussed in Section 3.6. Supercritical fluid technologies are being studied, although the technology is at an early stage. More developed methods include (1) precipitation and (2) emulsion as template.

Precipitation. Typically, the water-insoluble API is dissolved in an organic solvent, which is then mixed with an antisolvent, usually water. The addition of water causes a rapid supersaturation (nuclei formation) and growth of nanosized crystalline or amorphous drug. The limitation of this technique is that the drug needs to be soluble in at least one solvent, and the solvent needs to be miscible with the nonsolvent. This technology is available from DowPharma and BASF Pharma Solutions.

Emulsions as template. An emulsion is initially prepared comprising an organic solvent, or a mixture of solvents, loaded with the drug, which is dispersed in an aqueous phase containing suitable surfactants. The organic phase is then evaporated so the drug particles precipitate instantaneously to form a nanosuspension stabilized by surfactants. It is possible to control the particle size of the nanosuspension by controlling the size of the emulsion droplets. However, the possible use of hazardous solvents in this process raises safety and cost concerns.

A hybrid approach is also feasible: for example, Baxter’s Nanoedge^® technology employs both “bottom-up” and “top-down” approaches, through microprecipitation and also homogenization. Significant progress in nanocrystal preparation technology is being made, which should lead to substantially more products being brought to market in the near future.

3.3 SALT FORMATION

Salt formation, as described in Chapter 2 (Section 2.2.1), is the most common and effective method of increasing solubility and dissolution rates of acidic and basic drugs. In fact, more than 50% of the drugs currently listed in the USP are salt forms. The actual solubility of a salt, which is governed by the solubility products of the API and the counter salt, may not be much better than that of the free form. However, the dissolution rate is usually much faster, because of alterations in the microenvironmental pH. This phenomenon can be explained by considering that a weakly acidic drug is unionized in the stomach and therefore has a low dissolution rate. If the free acid is converted to the corresponding sodium or potassium salt, the strongly alkali sodium or potassium cations exert a neutralizing effect. Thus, in the immediate vicinity of the drug, the pH is raised to, for example, pH 5–6, instead of pH of 1–2 in the bulk medium of the stomach, thereby resulting in an alkaline microenvironment around the drug particle. This causes dissolution of the acidic drug in this localized region of higher pH, which gives rise to overall faster dissolution rates. When the dissolved drug diffuses away from the drug surface into the bulk of the gastric fluid where the pH is again lower, the free acid form may precipitate out. However, the precipitated free acid will be in the form of very fine, wetted, drug particles. These drug particles exhibit a very large total effective surface area in contact with the gastric fluids, much larger than would have been obtained if the free acid form of the drug had been administered. This increase in surface area results in an increased dissolution rate. Similarly, a strong acid salt of a weak base causes a localized drop in pH around the drug, which enhances the dissolution of weak bases.

An obvious limitation of this approach is that salt formation is limited to those APIs with at least one acidic or basic group. If the approach is feasible, sodium salts are most commonly used for acidic drugs and hydrochloride salts are most commonly used for basic drugs. However, this does not imply that these salts have necessarily the highest solubilizing potential; merely that the long history of their use means that there is correspondingly more information available about them. They also have low toxicity and a low molecular weight, thereby minimizing the overall mass of the dose as a salt form, compared to the free base/acid. It should be stressed though, that the most effective salt depends on the particular API under study—for optimal results, the choice of salt should be determined via a rigorous screening process.

Figure 3.3 shows an example of the effect of salts on solubility and dissolution rates of a basic compound, haloperidol, in 0.01 M HCl (Li et al. 2005). The fastest dissolution rate was observed for the mesylate salt, in accordance with its higher solubility than the hydrochloride or phosphate salts. The dissolution of the hydrochloride salt was suppressed because of the common-ion effect, i.e., release of hydrochloride ions was hindered because of the presence of the same ion in the testing solution (the experiment was carried out in 0.1 M HCl). The common-ion effect may be of particular relevance for the oral bioavailability for hydrochloride salts, as HCl is also present in the gastric fluids.

FIGURE 3.3 Dissolution profiles of haloperidol mesylate, hydrochloride, and phosphate at 37°C. The surface area of the test solids was 0.5 cm². The dissolution medium was 0.01 M HCl. (From Li, S., Doyle, P., Metz, S. et al.: Effect of chloride ion on dissolution of different salt forms of haloperidol, a model basic drug. J. Pharm. Sci. 2005. 94. 2224–2231. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)

3.4 COCRYSTALS

Cocrystals have received much attention recently as a novel approach for overcoming low-solubility problems. The strategy centers on cocrystallizing the API with a crystalline solid, designated the coformer. This may produce a cocrystal with more favorable physicochemical properties, in which the API and coformer are connected by noncovalent interactions. Crystal engineering facilitates optimal coformer association for each particular API. Although there are various combination patterns of drug and coformer molecules, the “synthon approach” is frequently used in the design of a cocrystal (Thakuria et al. 2013), whereby particular functional groups of the API interact with complementary functional groups of the conformers, to form “supramolecular synthon” cocrystals (Figure 3.4).

Cocrystals are differentiated from salts by assessing the degree of proton transfer. In general, cocrystal formation is expected for ΔpK_a < 3 and salt formation is expected for Δpk_a > 3. If one component is liquid at room temperature, then the crystals are designated solvates, whereas if both components are solids at room temperature, then the crystals are designated as cocrystals. Although solvates have the potential to enhance drug dissolution rate (for example, the solvated forms of spironolactone), hydrates generally exhibit slower dissolution rate relative to anhydrates. Also, they are often physically unstable, which can lead to desolvation during storage and possible crystallization into less soluble forms. High solvent levels may also cause toxicity problems. Cocrystals, in contrast, are typically produced by more rational design and so tend to be more stable.

As is the case for salt formation, the solubility of cocrystals is governed by the solubility product of the API and the coformer. Practically speaking, a coformer with high solubility tends to effectively increase cocrystal solubility. For carbamazepine, the solubility was increased by more than two orders of magnitude when it was combined with nicotinamide or glutamic acid, but the increase was only 2-fold with the aid of salicylic acid (Good and Rodríguez-Hornedo 2009).

On a laboratory scale, cocrystals can be prepared by either grinding or precipitation. The grinding method, notably solvent-assisted grinding, is believed to be better for estimating cocrystallization ability (Friscic et al. 2006), whereas the precipitation method is more suitable for screening. For industrial production, hot-melt extrusion may be used. Importantly, a recent Guidance for Industry issued by the FDA has specified that cocrystals need not be regarded as a novel API. This significantly reduces the regulatory hurdles needed for their licensing, which represents a distinct advantage for this approach.

FIGURE 3.4 Representative hydrogen-bonding synthons for forming cocrystals, showing the (a) carboxylic acid dimer synthon, (b) the amide dimer synthon, (c) the acid-pyridine heterosynthon, and (d) the acid-amide heterosynthon.

3.5 POLYMORPHS

Depending on the conditions (temperature, solvent, time) and method used for crystallization, the molecules in a crystal can arrange in different ways: either they may be packed differently in the crystal lattice or there may be differences in the orientation, or conformation, of the molecules at the lattice sites (Florence and Attwood 2011). Different crystalline forms of the same compound are called “polymorphs.” Although chemically identical, the different crystal lattices are at different free energy states. At a given temperature and pressure, only one of the crystalline forms is stable and the others are known as metastable forms. A metastable polymorph usually exhibits a greater aqueous solubility and dissolution rate, and thus greater absorption, than the stable polymorph.

Various physicochemical characteristics, including solubility and reactivity, can be correlated with their free energy difference, ΔG_A–B, according to the following equation:

$Δ G_{A - B} = R T ln (\frac{r_{A}}{r_{B}}) = R T ln (\frac{x_{A}}{x_{B}})$

(3.2)

where

r_A and r_B are the rates of chemical reaction of forms A and B, respectively, x_A and x_B are their solubilities, respectively

R is the gas constant

T is the temperature

The actual performance may or may not be predicted from this equation. For example, the dissolution rate may be governed by particle size rather than solubility. Nevertheless, thermodynamics is the basis for understanding various physicochemical characteristics.

A comparison of the melting enthalpy is a particularly useful method for determining the thermodynamic relationship between different polymorphic forms. Figure 3.5 shows a plot of the free energy of two polymorphs against temperature. The most stable form at ambient temperature is defined as Form I and the metastable form as Form II. For an enantiotropic system (Figure 3.5a), their free energies become equal at the transition temperature $T_{t}^{I - II}$ . However, the experimental transition temperature may be different from the theoretical one. In many cases, the transition is observed at a higher temperature, because of the high-energy barrier of the transformation. It is even possible that Form I melts without transformation. In this case, the melting enthalpy of Form I is usually larger than that of Form II, as indicated by the difference in the length of the arrows in Figure 3.5a. Because the enthalpy of Form I is lower than that of Form II at that temperature, the transition from Form I to Form II is an endothermic process. Thus, it may be possible to convert reversibly between the two polymorphs on heating and cooling (Kawakami 2007). Conversely, a monotropic system is shown in Figure 3.5b: the stability order remains unchanged below their melting temperatures. Metastable Form II may or may not cause polymorphic transition to Form I upon heating. If this occurs, the transition is an exothermic process and irreversible. Unless the transformation occurs, the metastable Form II melts with a smaller melting enthalpy than that of the stable Form I.

FIGURE 3.5 Free energy (G) and enthalpy (H)—temperature (T) diagrams of two crystal forms, which are in relationship of (a) enantiotropy or (b) monotropy. The superscripts I, II, I–II, and L represent Form I, Form II, Form I–II transition, and the liquid state, respectively. Form I is the stable form in the case of the monotropic relationship and the stable form at lower temperature in the case of the enantiotropic relationship. T_t and T_m are the transition and melting temperatures, respectively. ΔH_t and ΔH_m are the enthalpies of polymorphic transition and melting, respectively.

Most modern drugs show polymorphism, frequently being able to crystallize in three or more forms. The most stable polymorph has the lowest solubility and slowest dissolution rate. Therefore, selecting the metastable form represents a potential strategy in order to improve the solubility of poorly soluble drugs. A review of the literature concluded that solubility ratios were less than 2 in most cases, with an average value of 1.7 (Pudipeddi and Serajuddin 2005) suggesting that in many cases, solubility gains are small. However, in some cases, the free energy differences between polymorphs are large enough to significantly affect solubility and hence bioavailability. An example is chloramphenicol palmitate, where one polymorphic form was shown to attain approximately seven times greater blood levels in comparison to another (Aiguiar and Zelmer 1969).

In addition to solubility considerations, the issue of polymorphism has profound implications for the stability of a formulation and thus must be studied very carefully in this context. Under a given set of conditions, the polymorphic form with the lowest free energy will be the most stable, and other polymorphs will tend to transform into it. The rate of conversion is variable and is determined by the magnitude of the energy barrier between the two polymorphs. If there is a high-energy barrier and the crystal is stored at a low temperature, it can be expected that the conversion rate will be slow. Occasionally, the most stable polymorph appears only several years after the compound was first marketed. A case in point is the development of the antiretroviral drug, ritonavir. The most stable crystalline form appeared 2 years after the initial product launch and demonstrated solubility approximately 4- to 5-fold less than the original crystal form of the drug. The original soluble crystal form could no longer be generated, so that the newly emerged, more stable form required reformulation into gelcaps and tablets, rather than the original capsules, to ensure adequate oral bioavailability (Bauer et al. 2001). With such possible risks, it is perhaps understandable that pharmaceutical companies may be reluctant to employ the metastable form for the purpose of solubility improvement. Nevertheless, in some cases, it can be an excellent option for overcoming the solubility problem.

3.6 AMORPHOUS SOLID DISPERSIONS

Crystalline solids are packed in a regularly ordered, repeating pattern, whereas the corresponding amorphous form is characterized by a random arrangement of the molecules and an absence of long range, 3D, order. Because the amorphous form of a drug has no crystalline lattice, dissolution is more rapid, as no energy is required to break up the crystal lattice during the dissolution process; wettability is also typically better for the amorphous form. Therefore, the bioavailability of the amorphous form of an API is generally greater than that of the crystalline form: for example, the amorphous form of novobiocin is at least 10 times more soluble than the crystalline form (Florence and Attwood 2011).

Figure 3.6 shows both the ideal dissolution profile expected from the energy state of the amorphous form according to Equation 3.2 and also a typical profile of a real amorphous solid. The discrepancy is usually because soon after suspending an amorphous form in aqueous media, crystalline solids appear. Thus, the solution concentration does not reach the ideal amorphous solubility x_a but exhibits a peak at x_{a_real}, followed by a gradual decrease, until the concentration reaches the solubility of the crystal, x_c, after a sufficient period of time. Note that x_a is not supersaturation, but a solubility equilibrium for the amorphous form. However, once the crystalline drug appears, the solution with a solubility greater than the crystalline solubility can be regarded as being in a supersaturated state, relative to the crystalline solubility. This is the “spring-and-parachute” concept of amorphous dissolution and subsequent absorption: the formation of a supersaturated solution is “the spring,” to drive up solubility in the GI tract; after which supersaturation must be maintained long enough to drive drug absorption—“the parachute” phase (Brouwers et al. 2009). The “parachute” phase must be maintained for a long time to achieve an improvement in the oral absorption behavior.

FIGURE 3.6 Dissolution profile of the crystalline state, ideal profile of the amorphous state, and a typical profile of the real amorphous solid. The theoretical crystalline and amorphous solubilities are represented by x_c and x_a, respectively. The real amorphous solids exhibit a peak at x_{a_real}, followed by a gradual decrease in the concentration due to the appearance of crystalline solids.

The dissolution advantage of amorphous solids can be negated by crystallization of the amorphous solid in contact with the dissolution medium, as well as rapid crystallization of the supersaturated solution. Furthermore, there is a risk that the high-energy amorphous state will convert to the crystalline state with time, leading to a decrease in the dissolution rate with ageing. There are also difficulties associated with processing the amorphous material. In order to overcome these disadvantages, much research has been focused on developing ASDs, in which the amorphous form is associated with a hydrophilic polymeric carrier, as described next.

3.6.1 MISCIBILITY WITH EXCIPIENTS

The term “solid dispersion” is a general one, denoting any formulation in the solid state, in which the API is dispersed in an inert matrix. Typically, a hydrophobic API is dispersed in a hydrophilic matrix carrier. Although several classifications of solid dispersions exist (including simple eutectic mixtures, solid solutions, glass solutions, and glass suspensions), it is ASDs that have received the most interest in drug delivery and are discussed here. In an ASD, an amorphous drug is molecularly dispersed within a solid matrix. The matrix comprises a polymeric carrier, that is typically hydrophilic, amorphous, and has a high glass transition temperature (T_g). Polymers such as PVP, HPMC, and its derivatives are commonly used. Novel polymers have also been especially designed for this purpose, such as Soluplus^® (BASF), a polyvinyl caprolactam–polyvinyl acetate–PEG graft copolymer.

ASDs demonstrate enhanced solubility for a number of reasons: (1) the drug is present in the amorphous form and thus demonstrates more rapid dissolution in comparison to its crystalline counterpart (spring-and-parachute effect), (2) the drug is arranged within the carrier as a molecular dispersion, with a maximally reduced particle size, (3) the drug is intimately associated with a hydrophilic amorphous carrier, and (4) a recent important finding is that supersaturation can be achieved via the formation of nanoparticles, composed of the API and the polymeric excipients (Alonzo et al. 2010).

ASD formulations also have important advantages with respect to stability. ASDs significantly reduce the dangers of crystallization (of both the amorphous solid and the supersaturated solution) that can occur in contact with the dissolution medium, thus leading to the generation of supersaturated solutions that can persist for biologically relevant timeframes (Alonzo et al. 2010). Furthermore, as the drug is “locked” within the polymeric carrier in the solid state, its molecular motion is very low. Therefore ASDs also provide enhanced physical stability compared to the amorphous form alone, so that long-term storage stability is significantly improved.

In order to optimize the extension of parachute behavior, careful consideration of the excipient species and mixing ratio is required. Although a larger polymer/API ratio is preferred for physical stabilization and effective creation of supersaturated state, this also increases formulation volume. Thus, an optimum mixing ratio must be determined for each drug–polymer pairing. The free energy of mixing is described by the Flory–Huggins equation:

$\frac{Δ G_{i}}{k T} = ϕ_{d}^{i} \ln ϕ_{d}^{i} + \frac{ϕ_{p}^{i}}{N} \ln ϕ_{p}^{i} + χ ϕ_{d}^{i} ϕ_{p}^{i}$

(3.3)

where

ΔG_i, $ϕ_{d}^{i}$ , and $ϕ_{p}^{i}$ are the mixing free energy, drug fraction, and polymer fraction of phase i, respectively

χ is the interaction parameter between the drug and polymer, for which a value <0.5 indicates a miscible combination and a value greater indicates an immiscible combination

k and N are the Boltzmann constant and segment number of the polymer molecule, respectively

The overall mixing free energy ΔG can be obtained from

$Δ G = X_{d} Δ G_{d} + X_{p} Δ G_{p}$

(3.4)

where

subscripts d and p represent the drug-rich and polymer-rich phases, respectively

X is the fraction of each phase

Figure 3.7 shows examples of the phase diagram drawn by minimizing ΔG. A significant expansion of the two-phase region is observed with an increase in N. Thus, an increase in the molecular weight of a polymeric excipient can cause phase separation. Because the molecular weight of a drug molecule is usually in the range of 200–1000 Da, N is >100 in most cases, where the drug solubility rapidly decreases above χ = 1. Because high miscibility cannot be expected for a combination of a hydrophilic polymer and a poorly soluble drug, χ is expected to be >1. Even for χ = 1, phase separation is expected under drug-rich conditions, when N is >8 and the drug solubility in the matrix is almost constant at ca. 33% for N > 100. The expected solubility becomes only 15% when χ = 1.5 and is <10% when χ = 2.0. This theoretical calculation demonstrates why only a small amount of drug can be loaded into a stable solid dispersion.

FIGURE 3.7 Theoretical phase diagrams of solid dispersions composed of polymeric excipients and the active pharmaceutical ingredients, which were established on the basis of the Flory–Huggins theory. The miscible system and the system that induces phase separation are described as “1 phase” and “2 phases,” respectively.

3.6.2 STORAGE STABILITY

One of the significant issues that has hampered the development of ASDs until recently is their lack of physical stability. As outlined earlier, there is a risk that the amorphous state within the carrier will convert to the crystalline state, leading to a decrease in the supersaturation effect with ageing. X-ray powder diffraction (XRPD) is the most convenient method for evaluating whether a formulation is in the amorphous state. However, the XRPD technique cannot detect nuclei and small crystals that should enhance the crystallization. If the API and polymeric carrier are completely mixed, the formulation has only one T_g, which can be determined by DSC. A formulation with multiple T_gs indicates the existence of multiple phases. Although phase separation itself is not a fatal problem, the probability of crystallization of the API-rich phase is high.

Physical stability is governed by T_g of the system. A high T_g of the carrier increases the overall T_g of the mixture, resulting in a more stabilized amorphous form. Recent investigations have revealed that the amorphous form can be expected to be stable for 3 years at 25°C, if the system T_g is higher than 48°C (Kawakami et al. 2014). The absorption of moisture decreases the stability significantly, because molecular mobility is enhanced.

For crystalline formulations, accelerated studies for assuring chemical stability are based on the Arrhenius rule; however, the possibility of the crystallization of the amorphous form cannot be quantitatively predicted from accelerated studies. Furthermore, if phase separation or crystallization occurs at elevated temperatures, it may affect chemical stability. This unpredictability of the amorphous form stability leads to a prolongation of the drug developmental period. Currently, much research effort is directed at mechanisms to improve long-term stability predictions.

3.6.3 IMPROVEMENT IN ORAL ABSORPTION

In order to achieve an improvement in oral absorption, it is important to obtain a supersaturated state for a prolonged period (“spring and parachute”). The conventional dissolution test protocol utilizes sink conditions and focuses on the dissolution rate of the API, rather than providing an assessment of supersaturation. As such, it is not very effective in predicting the oral absorption enhancement, or if crystallization will occur in the GI tract.

Figure 3.8 compares three different griseofulvin (GF) formulations: a physical mixture, a solid dispersion containing both amorphous and crystalline GF, and an ASD. The ASD demonstrated the highest plasma concentration. The solid dispersion that contained both amorphous and crystalline GF displayed similar in vitro and in vivo profiles to that obtained for the physical mixture. Using the crystalline GF as a template, this can be explained by the immediate crystallization of the amorphous solid and/or dissolved solutes. This study demonstrates the importance of both complete amorphization of the API, and the maintenance of supersaturation, in order to improve oral absorption.

3.6.4 MANUFACTURE

In principle, most classes of material can be prepared in the amorphous state, if the rate at which they are solidified is faster than that at which their molecules can align themselves into a crystal lattice with 3D order. Small-scale production of solid dispersions can be achieved by the melt-quenching method, whereby the API is heated above its melting temperature, followed by rapid cooling. The required cooling rate depends on the glass-forming ability of the compound. Another simple method for obtaining amorphous solids is by precipitation, by adding a poor solvent to the API solution. Grinding, by supplying mechanical and thermal energy, can also convert a crystalline drug to the amorphous state. Although it is a simple procedure, it frequently leaves nuclei or small crystals, which can accelerate crystallization during storage.

FIGURE 3.8 (a) In vitro dissolution profiles and (b) plasma concentrations after oral administration of griseofulvin (GF) formulations to rats (7.5 mg/kg). A physical mixture (square), a solid dispersion that contains both amorphous and crystalline GF (triangle), and an ASD (circles) are shown. All of the formulations were prepared with Eudragit L-100 as an excipient. The Japanese Pharmacopeia-specified simulated intestinal solution of pH 6.8 (JP-2 solution) with 20 mM taurocholic acid was used as a medium in the dissolution test. (Reprinted with permission from Zhang, S. et al., Mol. Pharm., 8, 807. Copyright 2011 American Chemical Society.)

FIGURE 3.9 Hot-melt extrusion for the preparation of amorphous solid dispersions. (Courtesy of BASF SE, Ludwigshafen, Germany.)

A major problem with ASDs in the past, which significantly hampered progress in the field, was the difficulty in scaling up to large-scale industrial manufacturing. However, this problem has now been largely overcome due to the successful application of hot-melt extrusion technology, which can be a continuous process, in which equipment is available that allows solvent-free manufacture at temperatures above the relevant T_g (Figure 3.9). In this process, a crystalline API and an amorphous polymer are fed into the extruder, before being conveyed and exposed to shear stress. This transforms the API into its amorphous form, which is blended with the polymer and codispersed. Surfactants and other excipients may be added to aid in the extrusion process or to improve dissolution performance in vivo. The resulting solid (referred to as extrudate) is then pressed out and collected for further processing, e.g., into granules, spheres, powders, films, patches, and injections (Figure 3.9). Spray-drying is also frequently utilized for industrial production of ASDs.

3.6.5 COMMERCIAL EXAMPLES

ASDs can be prepared as tablets and as such offer significant advantages. For example, Kaletra^®, an HIV protease inhibitor, was initially marketed as a liquid self-emulsifying formulation. Patients were required to take six large soft capsules once daily (or three capsules twice daily); moreover, the capsules required refrigeration for storage. This formulation was replaced by ASD tablets, prepared using hot-melt technology. The number of the tablets required was reduced to four a day (or two tablets twice a day) and the tablets could be stored at room temperature.

Although there has been over 40 years of research in the area, the number of commercial products using ASDs is not extensive (Table 3.2). Research has been hindered by the manufacturing and stability issues described earlier. However, these issues have been largely addressed and continue to be improved upon, so it is expected that more commercial products based on ASDs will soon follow.

3.7 CYCLODEXTRINS

CDs are cyclic oligosaccharides, based on α-D-glucopyranose units, which can improve water solubility by complexation of the API within a cavity (Figure 3.10). The hydrophilic exterior surface of the CD molecules makes them water soluble, but the hydrophobic parts of central cavity provides a microenvironment for the housing of the hydrophobic parts of APIs. A range of CDs are available (α-CD, β-CD, γ-CD), with varying numbers of sugar residues and thus different-sized cavities, to accommodate guest APIs with various structures. More water-soluble CD derivatives are also available, including the hydroxypropyl derivatives of β-CD and γ-CD and the randomly methylated β-CD (RMβCD).

TABLE 3.2
Examples of Commercial Products Using Amorphous Solid Dispersions

Commercial Preparation	Dmg
Accolate^®	Zafirlukast
Accupril^®	Quinapril hydrochloride
Ceftin^®	Cefuroxime axetil
Cesamet^®	Nabilone
Certican^®	Everolimus
Crestor^®	Rosuvastatin calcium
Gris-PEG^®	Griseofulvin
Intelence^®	Etravirine
Isoptin^®	Verapamil
Kaletra^®	Lopinavir/ritonavir
Nivadil^®	Nilvadipine
Prograf^®	Tacrolimus
Rezulin^®	Troglitazone
Sporanox	Itraconazole
Zelboraf^®	Vemurafenib

CDs as solubility enhancers have been used in the pharmaceutical industry for many years: the first CD product was introduced in Japan in 1976 (Prostarmon E^®) and comprised the prostaglandin E2 solubilized by β-CD as the molecular inclusion complex. The first U.S.-approved product was the antifungal itraconazole in a hydroxypropyl-β-CD complex (Sporanox^® oral solution) and was introduced in 1984. Currently, 35 different drugs are marketed worldwide as either solid or solution-based CD complex formulations.

Biocompatible CDs are obviously safer and less toxic than harsh organic solvents for improving water solubility. They are also less likely to have problems of precipitation in vivo in contact with aqueous body fluids. This is because CDs solubilize compounds as a linear function of their concentration if they form 1:1 complexes, so that in contact with aqueous fluids in vivo, both the drug and CD concentration are reduced in a linear manner. In contrast, organic solvents solubilize solutes as a log function of their concentration, so that precipitation is more likely to occur with the rapid dilution that occurs in contact with the aqueous environment.

3.8 MICELLAR SOLUBILIZATION OF DRUGS

3.8.1 SURFACTANT MICELLES

Surfactants are amphiphilic molecules typically comprising a long-chain hydrocarbon tail and a head group that can be either (1) anionic, e.g., sodium dodecyl sulfate; (2) cationic, e.g., dodecyltrimethylammonium bromide; or (3) nonionic, e.g., n-dodecyl tetra(ethylene oxide). At low concentrations, surfactants are widely used as formulation excipients. For solid dosage forms, the addition of even a small amount of surfactant helps improve solubility, because it can improve wetting properties and aid in the rapid disintegration of the dosage form. Surfactants are also used as stabilizers in many other delivery systems, such as emulsions, microemulsions, nanocrystals, and ASDs. Other complementary effects of surfactants may play a role in enhancing oral bioavailability. For example, nonionic surfactants such as TPGS present in a formulation can inhibit P-glycoprotein efflux pumps. Furthermore, many surfactants function as absorption enhancers, promoting both transcellular and paracellular transport pathways across the GI epithelium (see also Chapter 7).

FIGURE 3.10 Schematic representation of formation of a cyclodextrin–drug 1:1 complex. (Courtesy of Pierre Fabre Medicament—Supercritical Fluids Division.)

Their role as solubilizers is due to their self-aggregation properties: when surfactant molecules are dissolved in water at concentrations above the critical micelle concentration (CMC), they form colloidal-sized aggregates known as micelles, in which the hydrophobic portions are driven inward, to form a hydrophobic core, while the hydrophilic portions face outward, toward the water. Surfactant micelles increase the solubility of poorly soluble substances in water, because the nonpolar drug molecules are solubilized within the hydrophobic micelle core. Polar molecules (or polar portions of a drug molecule) will be adsorbed on the micelle surface, and substances with intermediate polarity are distributed along surfactant molecules in certain intermediate positions (Florence and Attwood 2011). The capacity of surfactants to solubilize drugs depends on various factors, including the physicochemical nature of both the drug and surfactant, the temperature, pH, etc.

However, surfactant micelles are not static aggregates: they dissociate, regroup, and reassociate rapidly (hence, they are often referred to as “association colloids”). The solubilizing capacity of the surfactant can be lost on dilution with aqueous fluids in vivo. Micellar collapse can lead to drug precipitation, for example, in the lumen of the GI tract following oral administration, or in contact with the blood after injection. Furthermore, the solubilization capacity for poorly soluble drugs in surfactant micelles is usually less than 20 mg/g of surfactant, meaning that gram amounts of surfactant are usually required for complete solubilization, which is not often realistic practically. At high concentrations, surfactants may possibly cause damage to the GI epithelium, disrupting proteins in the plasma membrane (see also Chapter 7). If given intravenously, surfactants at high concentrations can cause anaphylactic reactions and other toxicity issues.

3.8.2 POLYMERIC MICELLES

The amphiphilic block copolymers are a newer class of surfactants, which are able to form stable micelles, known as polymeric micelles, at low CMC values (Torchilin 2004). The amphiphilic block copolymers typically consist of (1) a hydrophobic polyester block (for example, polylactic acid [PLA] or poly(lactic-co-glycolic acid) [PLGA]) and (2) a hydrophilic block comprising PEG or poly(ethylene oxide). When the length of the hydrophilic block exceeds the length of the hydrophobic block, these copolymers can form spherical micelles in aqueous solution (see also Chapter 1, Figure 1.14). Again, the hydrophobic blocks form the micellar core, which can accommodate a poorly soluble drug; the hydrophilic blocks form the outer shell. Polymeric micelles are being extensively investigated for drug delivery applications, in particular for parenteral administration. For example, Genexol-PM^® is a polymeric micelle formulation of paclitaxel for the treatment of breast cancer. It is composed of a low-molecular-weight amphiphilic diblock copolymer, monomethoxy PEG-block-poly(D,L-lactide). In addition to being able to solubilize poorly soluble drugs at low CMC values, the stability and outer hydrophilic layer of polymeric micelles promote long circulation times in the blood, which can allow time for their accumulation at sites of inflammation and infection (attributed to the enhanced permeability and retention [EPR] effect). Furthermore, they can be actively targeted to the site of action, by the attachment of a specific targeting vector to the outer surface. These drug delivery and targeting aspects of polymeric micelles are described further in Chapter 5 (Section 5.5.3).

3.9 OILS, EMULSIONS, AND COLLOIDAL CARRIERS

3.9.1 OILS AND COARSE EMULSIONS

A simple approach to enhancing the solubility of “grease ball” molecules is to dissolve the hydrophobic drug in an oily liquid; the solubilized drug can then be filled into capsules for oral delivery. Medium-chain mono-, di-, or triglycerides and their esters are convenient solvent choices, because of their large solubilization capacity and good compatibility with capsules. Oils can also be used to prepare oil-in-water (O/W) emulsions, to solubilize hydrophobic drugs within the oil phase. These systems are thermodynamically unstable, due to the large interfacial energy between the oil and water phase.

After oral administration, oils and coarse emulsions are subjected to the physiologically complex processes of digestion and absorption that exist for dietary lipids within the GI tract. Oil-based delivery systems initially form coarse droplets within the aqueous GI fluids, which then require the detergent action of bile salts to emulsify them into smaller, stabilized droplets. These smaller droplets are then subjected to the action of pancreatic lipase, which digests the oil into free fatty acids and monoglycerides, finally liberating the drug. The hydrophobic drug, with the free fatty acids and monoglycerides of the digested oil carrier, in addition to cholesterol and lecithin, are all subsequently incorporated into bile salt micelles. These micelles serve as carriers, to shuttle the hydrophobic API through the aqueous GI contents, to reach the absorbing surface of the enterocyte. The secretion and activity of bile salts demonstrate high inter- and intravariability and are profoundly affected by the fasting state of the patient, thus introducing significant variability into the drug absorption process for these drug delivery systems (DDS).

3.9.2 SELF-MICROEMULSIFYING DRUG DELIVERY SYSTEMS

A promising advance in emulsion technology is the development of self-emulsifying DDS (self-emulsifying drug delivery systems [SEDDS]), which comprise physically stable, isotropic mixtures of oils, surfactants, solvents, and cosolvents/surfactants, in very specific combinations, that require careful selection (Hauss 2007). Typical components may include

• Oils: Such as peanut oil and medium-chain triglycerides (e.g., Neobee^® M5)

• Emulsifiers: Surfactants such as Labrafac^® CM-10 and polyglycolyzed glycerides with varying fatty acid and PEG chain lengths

• Cosolvents: Such as propylene glycol and Transcutol^®

SEDDS can be orally administered in soft or hard gelatin capsules. On dilution in the aqueous GI fluids and facilitated by the gentle agitation of the GI contents due to peristaltic activity, they spontaneously self-form into very fine, relatively stable O/W emulsions, with lipid droplets size of about 100–200 nm (SEDDS), or less than 100 nm for self-microemulsifying drug delivery systems (SMEDDS). SMEDDS thus differ from coarse emulsions in that they are a thermodynamically equilibrium solution and as such, form spontaneously, without energy input. A phase diagram may be used to obtain the optimum composition of the formulation, to promote optimal microemulsification.

Oral absorption is facilitated by the rapid release of drug from the high surface area of the small lipid droplets. SMEDDS do not require bile salts and other digestive processes for their digestion, thereby minimizing inter- and intrasubject variations. For SMEDDS, there is also the possibility that due to the lipidic nature of the delivery system, lymphatic absorption via the intestinal lacteals, or the intestinal Peyer’s patches, is enhanced. Drugs absorbed via the lymphatics avoid the “first-pass” effects of the liver (see Chapter 7), which represents a significant advantage for the oral absorption of enzymatically labile APIs.

Supersaturatable SEDDS (S-SEDDS) are an extension of the original SEDDS, which incorporate less surfactant than original system, in order to avoid the adverse effects of surfactants in the GI tract. There is a risk that with less surfactant in the formulation, the system may precipitate out on dilution with the GI fluids. However, the surfactant concentration can safely be reduced by incorporating a polymeric precipitation inhibitor (PPI), most commonly HPMC. The PPI prevents precipitation of the system in the GI tract and also maintains a supersaturated state of the drug for extended periods, thereby providing a strong driving force for the absorption process. Figure 3.11 shows an example of the effect of supersaturation in a study of various self-emulsifying formulations containing ca. 60 mg/g paclitaxel (Gao et al. 2003). As shown in Figure 3.11a, when the formulation without HPMC was diluted 50-fold using simulated gastric fluid, precipitation occurred in 10 minutes and the paclitaxel concentration decreased to ca. 0.1 mg/mL, where the equilibrium solubility was expected to be 0.02 mg/mL. In contrast, the paclitaxel concentration from the formulation with HPMC was greater than 0.9 mg/mL at 10 minutes, and a high level of supersaturation was maintained for more than 2 hours. Oral absorption was greatly enhanced by the formulation with HPMC, as shown in Figure 3.11b, which can be explained by the maintenance of a high concentration of dissolved paclitaxel in the small intestine. Thus, addition of a PPI to self-emulsifying formulations appears to be a very useful approach. However, surfactant and polymer molecules usually have strong interactions and may form various types of complexes, which also require attention during the design of the formulation.

SMEDDS can be incorporated into soft capsules for oral administration and examples now on the market include Agenerase^® (amprenavir), Aptivus^® (tipranavir), Fortovase^® (saquinavir), Kaletra^® (lopinavir/ritonavir), Neoral^® (cyclosporine A), Rapamune (sirolimus), and Xtandi^® (enzalutamide). Neoral^® is a SMEDDS reformulation of Sandimmune^®, which was a SEDDS introduced in 1994. Neoral^® emulsifies spontaneously into a microemulsion with a particle size smaller than 100 nm, which increases the bioavailability nearly by a factor 2 over the original emulsion formulation. In addition, Neoral^® shows a much faster onset of action, a reduced inter-/intrasubject variability and a much lower impact of food intake on cyclosporin pharmacokinetics.

FIGURE 3.11 (a) Dissolution behavior of two self-emulsifying formulations of paclitaxel. A 50-fold dilution in simulated gastric fluid at 37°C was performed. Paclitaxel concentration is shown after dilution of a formulation containing hydroxypropyl methylcellulose (HPMC) (circles), or without HPMC (triangles). (b) An oral absorption study of paclitaxel formulations in fasted rats, showing plasma paclitaxel concentrations after dosing with a formulation containing HPMC (circles), without HPMC (triangles), or a control formulation of Taxol^® (squares). (From Gao, P., Rush, B.D., Pfund, W.P. et al.: Development of a supersaturatable SEDDS (S-SEDDS) formulation of paclitaxel with improved oral bioavailability. J. Pharm. Sci. 2003. 92. 2386–2398. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)

3.9.3 OTHER COLLOIDAL CARRIERS

As described in Chapter 5, a wide variety of micro- and nanoparticulate DDS have been developed, particularly for the parenteral route of delivery (see also Chapter 5, Figure 5.1). Such DDS include liposomes, niosomes, lipoprotein carriers, polymeric micro- and nanoparticles, and dendrimers. Many of these systems offer the advantage of increasing the solubility of an API. For example, hydrophobic APIs can be associated with the lipid bilayers of a liposomal DDS (see Chapter 5, Figure 5.9). Hydrophobic drugs can also be accommodated within the hydrophobic core region of micro- and nanoparticles and other colloidal carriers. As these DDS are predominantly investigated for their drug delivery and targeting purposes, rather than for their solubilization potential, the reader is referred to Chapter 5 for further information on these other colloidal carriers.

3.10 CONCLUSIONS

Enhancing the water solubility of poorly soluble drugs continues to be a pressing concern in the pharmaceutical industry, with important implications for the drug discovery process, as well as for the formulation, bioavailability, and therapeutic efficacy of poorly soluble APIs. There are now many different ways to improve drug solubility, as outlined in this chapter. However, there is no single all-encompassing solution to this problem, and each approach has associated advantages and limitations. Furthermore, in considering which approach to use, it is important that each potential drug candidate is considered on an individual basis. A number of interrelated factors must be taken into account, including the following:

• The nature of the API: What type of API needs to be solubilized? Is it a grease ball or a brick dust molecule? What solubilizes the API? Can it dissolve in an oily/lipidic medium, or does it need a more complex blend of oils, surfactants, and cosolvents? How stable is the API and can it withstand harsh industrial processes?

• The delivery route: Parenteral, oral, etc.

• Therapeutic issues: The disease, the desired therapeutic outcome, the dose, and the duration of drug administration.

• Industrial issues: The availability of backup candidates, the developmental timeline, the available resources for development and manufacture, and the relevant regulatory issues.

Whatever method for solubility improvement is chosen, meticulous characterization and stability analysis are required, including the establishment of rigorous protocols to predict physical stability. Stability will need to be ascertained both in vitro and also in vivo. For the oral route, this will require a study of the stability and performance of the formulation within the highly variable and complex GI milieu.

The field has seen recent novel and exciting advances to improving drug solubility, such as innovative lipid delivery systems like SMEDDS, the success of NanoCrystal^® technology, and the reemergence of ASDs as a viable solubilization technology platform. Techniques such as hot-melt extrusion have allowed successful scale-up manufacturing of processes that were hitherto confined to the laboratory setting. There is currently intense research focus in this area, which is expected to yield further improvements in current methods, as well as the development of new approaches.

REFERENCES

Aiguiar, A.J. and J.E. Zelmer. 1969. Dissolution behaviour of polymorphs of chloramphenicol palmitate and mefenamic acid. Journal of Pharmaceutical Science 58:983–987.

Alonzo, D.E., G.G.Z. Zhang, D. Zhou et al. 2010. Understanding the behavior of amorphous pharmaceutical systems during dissolution. Pharmaceutical Research 27:608–618.

Amidon, G.L., H. Lennernä, V.P. Shah et al. 1995. A theoretical basis for a biopharmaceutic drug classification: The correlation of in vitro drug product dissolution and in vivo bioavailability. Pharmaceutical Research 12(3):413–420.

Bauer, J., S. Spanton, R. Henry et al. 2001. Ritonavir: An extraordinary example of conformational polymorphism. Pharmaceutical Research 18:859–866.

Bergström, C.A.S., C.M. Wassvik, K. Johansson et al. 2007. Poorly soluble marketed drugs display solvation limited solubility. Journal of Medicinal Chemistry 50:5858–5862.

Brouwers, J., M.E. Brewster, and P. Augustijns. 2009. Supersaturating drug delivery systems: The answer to solubility-limited oral bioavailability? Journal of Pharmaceutical Science 98:2549–2572.

Florence, A.T. and D. Attwood. 2011. Physicochemical Principles of Pharmacy, 5th ed. London, U.K.: Pharmaceutical Press.

Friscic, T., A.V. Trask, W. Jones et al. 2006. Screening for inclusion compounds and systematic construction of three-component solids by liquid-assisted grinding. Angewandte Chemie International Edition 45:7546–7550.

Gao, P., B.D. Rush, W.P. Pfund et al. 2003. Development of a supersaturatable SEDDS (S-SEDDS) formulation of paclitaxel with improved oral bioavailability. Journal of Pharmaceutical Science 92:2386–2398.

Good, D.J. and N. Rodríguez-Hornedo. 2009. Solubility advantage of pharmaceutical cocrystals. Crystal Growth & Design 9:2252–2264.

Hauss, D.J. 2007. Oral lipid-based formulations. Advanced Drug Delivery Reviews 59(7):667–676.

Kawakami, K. 2007. Reversibility of enantiotropically-related polymorphic transformations from a practical viewpoint: Thermal analysis of kinetically reversible/irreversible polymorphic transformations. Journal of Pharmaceutical Science 96:982–989.

Kawakami, K., T. Harada, K. Miura et al. 2014. Relationship between crystallization tendencies during cooling from melt and isothermal storage: Toward a general understanding of physical stability of pharmaceutical glasses. Molecular Pharmaceutics 11:1835–1843.

Li, S., P. Doyle, S. Metz et al. 2005. Effect of chloride ion on dissolution of different salt forms of haloperidol, a model basic drug. Journal of Pharmaceutical Science 94:2224–2231.

Merisko-Liversidge, E. and G.G. Liversidge. 2011. Nanosizing for oral and parental drug delivery: A perspective on formulating poorly-water soluble compounds using wet media milling technology. Advanced Drug Delivery Review 63(6):427–440.

Neslihan, R. and S. Benita. 2004. Self-emulsifying drug delivery systems (SEDDS) for improved oral delivery of lipophilic drugs. Biomedicine & Pharmacotherapy 58(3):173–182.

Pudipeddi, M. and A.T.M. Serajuddin. 2005. Trends in solubility of polymorphs. Journal of Pharmaceutical Science 94:929–939.

Thakuria, R., A. Delori, W. Jones et al. 2013. Pharmaceutical cocrystals and poorly soluble drugs. International Journal of Pharmacy. 453:101–125.

Torchilin, V.P. 2004. Targeted polymeric micelles for delivery of poorly soluble drugs. Cellular and Molecular Life Sciences 61(19–20):2549–2559.

Zhang, S., K. Kawakami, M. Yamamoto et al. 2011. Coaxial electrospray formulations for improving oral absorption of a poorly water-soluble drug. Molecular Pharmaceutics 8:807–813.

FURTHER READING

Brewster, M.E. and T. Loftsson. 2007. Cyclodextrins as pharmaceutical solubilizers. Advanced Drug Delivery Review 59:645–666.

Burger, A. and R. Ramberger. 1979. On the polymorphism of pharmaceuticals and other molecular crystals. I. Theory of thermodynamic rules. Mikrochimica Acta [Wien] II:259–271.

Kawakami, K. 2012. Modification of physicochemical characteristics of active pharmaceutical ingredients and application of supersaturatable dosage forms for improving bioavailability of poorly absorbed drugs. Advanced Drug Delivery Review 64:480–495.

Kesisoglou, F., S. Panmai, and Y. Wu. 2007. Nanosizing—Oral formulation development and biopharmaceutical evaluation. Advanced Drug Delivery Review 59:631–644.

Lawrence, M.J. and G.D. Rees. 2000. Microemulsion-based media as novel drug delivery systems. Advanced Drug Delivery Review 45:89–121.

Porter, C.J.H., C.W. Pouton, J.F. Cuine et al. 2008. Enhancing Intestinal drug solubilisation using lipid-based delivery systems. Advanced Drug Delivery Review 60:673–691.

Serajuddin, A.T.M. 1999. Solid dispersion of poorly water-soluble drugs: Early promises, subsequent problems, and recent breakthroughs. Journal of Pharmaceutical Science 88:1058–1066.

Strickley, R.G. 2004. Solubilizing excipients in oral and injectable formulations. Pharmaceutical Research 21:201–230.

Thakuria, R., A. Delori, W. Jones et al. 2013. Pharmaceutical cocrystals and poorly soluble drugs. International Journal of Pharmacy 453:101–125.

Williams, H.D., N.L. Trevaskis, S.A. Charman et al. 2013. Strategies to address low drug solubility in discovery and development. Pharmacological Reviews 65:315–499.