Advanced Drug Delivery Reviews 63 (2011) 456–469
`
`Contents lists available at ScienceDirect
`
`Advanced Drug Delivery Reviews
`
`j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / a d d r
`
`Physical and chemical stability of drug nanoparticles☆
`Libo Wu, Jian Zhang, Wiwik Watanabe ⁎
`MAP Pharmaceuticals, Inc. 2400 Bayshore Parkway, Mountain View, CA 94043, USA
`
`a r t i c l e
`
`i n f o
`
`a b s t r a c t
`
`As nano-sizing is becoming a more common approach for pharmaceutical product development, researchers
`are taking advantage of the unique inherent properties of nanoparticles for a wide variety of applications. This
`article reviews the physical and chemical stability of drug nanoparticles, including their mechanisms and
`corresponding characterization techniques. A few common strategies to overcome stability issues are also
`discussed.
`
`Published by Elsevier B.V.
`
`Article history:
`Received 12 August 2010
`Accepted 2 February 2011
`Available online 21 February 2011
`
`Keywords:
`Drug nanoparticles
`Nanosuspensions
`Physical stability
`Chemical stability
`Stabilizer
`
`Contents
`
`1.
`2.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Introduction .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Stability of drug nanoparticles .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`2.1.
`Effect of dosage form on stability .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`2.2.
`General stability issues related to nanosuspensions
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`2.2.1.
`Sedimentation or creaming .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`2.2.2.
`Agglomeration .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`2.2.3.
`Crystal growth .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`2.2.4.
`Change of crystalline state .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`2.2.5.
`Stability issues with solidification process of nanosuspensions .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`2.2.6.
`Chemical stability .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Additional stability issues relate to large biomolecules .
`.
`.
`.
`.
`.
`.
`.
`.
`2.3.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`Characterizing stability of drug nanoparticles and nanoparticle formulations .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`3.1.
`Particle size, size distribution and morphology.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`3.2.
`Sedimentation/creaming .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`3.3.
`Particle surface charge .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`3.4.
`Crystalline state .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`3.5.
`Chemical stability .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`3.6.
`Additional techniques for assessing large biomolecule nanoparticle and formulation stability
`Recommendations of general strategies for enhancing stability of nanoparticle formulations
`.
`.
`.
`4.
`Conclusions .
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`5.
`References
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`3.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`.
`
`456
`457
`457
`457
`457
`458
`463
`463
`464
`464
`464
`465
`465
`466
`466
`466
`466
`466
`467
`467
`467
`
`☆ This review is part of the Advanced Drug Delivery Reviews theme issue on
`“Nanodrug Particles and Nanoformulations for Drug Delivery”.
`⁎ Corresponding author. Tel.: +1 650 386 8193; fax: +1 650 386 3100.
`E-mail address: wwatanabe@mappharma.com (W. Watanabe).
`
`0169-409X/$ – see front matter. Published by Elsevier B.V.
`doi:10.1016/j.addr.2011.02.001
`
`1. Introduction
`
`With significant attention focused on nanoscience and nanotech-
`nology in recent years, nanomaterial-based drug delivery has been
`propelled to the forefront by researchers from both academia and
`industry [1–3]. Various nano-structured materials were produced and
`applied to drug delivery such as nanoparticles [4], nanocapsules [5],
`
`

`

`L. Wu et al. / Advanced Drug Delivery Reviews 63 (2011) 456–469
`
`457
`
`nanotubes [6], micelles [7], microemulsions [8] and liposomes [9]. In
`general, the term “nanoparticles” refers to particles with sizes
`between 1 and 100 nm. However, submicron particles are also
`commonly referred as nanoparticles in the field of pharmaceutics
`and medicine [10–14]. Nanoparticles are categorized as nanocrystals
`[10], polymeric [15], liposomal [9] and solid lipid nanoparticles (SLN)
`[16] depending on their composition, function and morphology. Given
`the extensive available literature reviews on SLN, polymeric and
`liposomal nanoparticles [4,9,15–18], this article will focus only on
`nanocrystals (pure drug nanoparticles).
`The unique nano-scale structure of nanoparticles provides signif-
`icant increases in surface area to volume ratio which results in notably
`different behavior, both in-vitro and in-vivo, as compared to the
`traditional microparticles [10–12]. Consequently, drug nanocrystals
`have been extensively used in a variety of dosage forms for different
`purposes [10,11,14,19,20], such as improving the oral bioavailability
`of poorly water-soluble drugs by utilizing enhanced solubility and
`dissolution rate of nanoparticles [21–23]. In the field of pulmonary
`drug delivery, the nanoparticles are able to deliver the drugs into the
`deep lungs, which is of great importance for systemically absorbed
`drugs [11,14]. In addition, injection of poorly water-soluble nanosus-
`pension drugs is an emerging and rapidly growing field that has
`drawn increasing attention due to its benefits in reducing toxicity and
`increasing drug efficacy through elimination of co-solvent in the
`formulation [10,20].
`Despite the advantages of drug nanocrystals, they present various
`drawbacks including complex manufacturing [12,24–26], nanotoxi-
`city [27] and stability issues [10,19,20]. Stability is one of the critical
`aspects in ensuring safety and efficacy of drug products.
`In
`intravenously administered nanosuspensions, for example, formation
`of larger particles (N5 μm) could lead to capillary blockade and
`embolism [20], and thus drug particle size and size distribution need
`to be closely monitored during storage. The stability issues of drug
`nanoparticles could arise during manufacturing, storage and shipping.
`For instance, the high pressure or temperature produced during
`manufacturing can cause crystallinity change to the drug particles
`[12,26,28]. Storage and shipping of the drug products may also bring
`about a variety of stability problems such as sedimentation,
`agglomeration and crystal growth [29–31]. Therefore, stability issues
`associated with drug nanocrystals deserve significant attention during
`pharmaceutical product development. This article reviews existing
`literature on drug nanoparticle stability, including theory/mecha-
`nisms, methods used to tackle the stability problems and character-
`ization techniques, and provides recommendations to improve the
`current practices. Since the stability issues related to nanoparticle dry
`powders are usually trivial, this review will only focus on stability of
`nanosuspensions (drug nanoparticles dispersed in a liquid medium).
`
`2. Stability of drug nanoparticles
`
`2.1. Effect of dosage form on stability
`
`The unique characteristics of drug nanoparticles have enabled
`their extensive application in various dosage forms including oral,
`parenteral, ocular, pulmonary, dermal and other specialized delivery
`systems [10,11,13,20,32]. Although different dosage forms may share
`some common stability issues, such as sedimentation, particle
`agglomeration or crystal growth, their effects on drug products are
`quite different. For instance, particle agglomeration could be a major
`issue in pulmonary drug delivery since it affects deposition amount/
`site, and thus drug efficacy. On the other hand, agglomeration in
`intravenous formulations can cause blood capillary blockage and
`obstruct blood flow. Moreover, the selection of stabilizers is also
`closely related to dispersion medium, dosage form and strictly
`governed by FDA regulations. To date, there is a wide variety of
`
`choices on the approved stabilizers for oral dosage form whereas the
`excipients allowed for inhalation are very limited [33].
`Drug nanoparticles exist in the final drug products either in dry
`powder or suspension form. Examples of the dry powder form include
`the dry powder inhaler, lyophilized powder for injection and oral
`tablets or capsules. Solid dosage forms usually have good storage
`stability profiles, which is why a common strategy to enhance
`nanosuspension stability is to transform the suspension into solid
`form [19,25]. Most of the reported stability concerns arise from
`nanosuspensions in which the drug nanoparticles are dispersed in a
`medium with or without stabilizers.
`In addition, mechanisms
`involved in the stability of small and large biomolecule formulations
`are different due to their molecular structure differences. A small
`molecule drug is defined as a low molecular weight non-polymeric
`organic compound while large biomolecules refer to large bioactive
`molecules such as protein/peptide. One of the major issues with
`protein/peptide stability is to maintain the 3-dimensional molecular
`conformation, such as secondary and tertiary structure in order to
`keep their biological activities [34,35], whereas there is no such
`concern for small organic molecules.
`
`2.2. General stability issues related to nanosuspensions
`
`Stability issues associated with nanosuspensions have been widely
`investigated and can be categorized as physical and chemical stability.
`The common physical stability issues include sedimentation/cream-
`ing, agglomeration, crystal growth and change of crystallinity state.
`
`2.2.1. Sedimentation or creaming
`Drug particles can either settle down or cream up in the
`formulation medium depending on their density relative to the
`medium. The sedimentation rate is described by Stokes' law [36,37]
`which indicates the important role of particle size, medium viscosity
`and density difference between medium and dispersed phase in
`determining the sedimentation rate. Decreasing particle size is the
`most common strategy used to reduce particle settling. Matching drug
`particles density with medium or increasing medium viscosity are the
`other widely used approaches to alleviate sedimentation problems
`[37,38]. Fig. 1 shows different sedimentation types that occur in
`suspension formulations.
`In a deflocculated suspension (Fig. 1a), particles settle indepen-
`dently as small size entities resulting in a slow sedimentation rate.
`However, densely packed sediment, known as caking [39], is usually
`formed due to the pressure applied on each individual particle. This
`sediment is very difficult to be re-dispersed by agitation [36,37,39]
`and would be detrimental to the drug products performance. In the
`flocculated suspension (Fig. 1b), the agglomerated particles settle as
`loose aggregates instead of as individual particles [36,37]. The loose
`aggregates have a larger size compared to the single particle, and thus
`higher sedimentation rate. The loose structure of the rapidly settling
`flocs contains a significant amount of entrapped medium and this
`structure is preserved in the sediment. The final flocculation volume is
`therefore relatively large and the flocs can be easily broken and re-
`dispersed by simple agitation. K.P.
`Johnston et al. [40,41] have
`recently attempted to achieve stable nanosuspensions via a novel
`design of flocs structure called “open flocs”, as illustrated in Fig. 1c.
`Thin film freezing was used to produce BSA nanorods with aspect
`ratio of approximately 24. These BSA nanorods were found to be
`highly stable when dispersed into hydrofluoroalkane (HFA) propel-
`lant, with no apparent sedimentation observed for 1 year. Due to the
`high aspect ratio of BSA nanorods and relatively strong attractive Van
`der Waals (VDW) forces at the contact sites between the particles,
`primary nanorods were locked together rapidly as an open structure
`upon addition of HFA, inhibiting collapse of the flocs [41]. The low-
`density open flocs structure was then filled with liquid HFA medium,
`preventing particle settling. Similar results were shown using needle
`
`

`

`458
`
`L. Wu et al. / Advanced Drug Delivery Reviews 63 (2011) 456–469
`
`considerations, selection of stabilizers is based on their ability to
`provide wetting to surface of the particles and offer a barrier to
`prevent nanoparticles from agglomeration [13,19].
`There are two main mechanisms through which colloidal suspen-
`sions can be stabilized in both aqueous and non-aqueous medium, i.e.
`electrostatic repulsion and steric stabilization [10,36,37]. These two
`mechanisms can be achieved by adding ionic and non-ionic stabilizers
`into the medium, respectively. Stabilization from electrostatic
`repulsion can be described by the classic Derjaguin–Landau–
`Verwey–Overbeek (DLVO) theory [43,44]. This theory mainly applies
`to aqueous suspension while its application in non-aqueous medium
`is still not well-understood [33]. The DLVO theory assumes that the
`forces acting on the colloidal particles in a medium include repulsive
`electrostatic forces and attractive VDW forces. The repulsive forces are
`originated from the overlapping of electrical double layer (EDL)
`surrounding the particles in the medium, and thus preventing
`colloidal agglomeration. The EDL consist of two layers: (i) stern
`layer composed of counter ions attracted toward the particle surface
`to maintain electrical neutrality of the system and (ii) Gouy layer
`which is essentially a diffusion layer of ions (Fig. 2).
`The total potential energy (VT) of particle–particle interaction is a
`sum of repulsion potential (VR) generated from electric double layers
`and attraction potential (VA) from the VDW forces. VA is determined
`by the Hamaker constant, particle size and inter-particulate distance
`while VR depends on particle size, distance between the particles, zeta
`potential, ion concentration and dielectric constant of the medium. VR
`is extremely sensitive to ion concentration in the medium. As the ion
`strength is increased in the medium, the thickness of EDL decreases
`due to screening of the surface charge [36,37]. This causes decrease in
`VR, increasing the susceptibility of the dispersed particles to form
`aggregates. Zeta potential (ZP) is electric potential at the shear plane
`which is the boundary of the surrounding liquid layer attached to the
`moving particles in the medium. ZP is a key parameter widely used to
`predict suspension stability. The higher the ZP, the more stable the
`suspension is.
`In the case of steric stabilization, amphiphilic non-ionic stabilizers
`are usually utilized to provide steric stabilization which is dominated
`by solvation effect. As the non-ionic stabilizers are introduced into
`nanosuspensions, they are absorbed onto the drug particles through
`an anchor segment that strongly interacts with the dispersed
`particles, while the other well-solvated tail segment extends into
`the bulk medium (Fig. 3).
`As two colloidal particles approach each other, the stabilizing
`segments may interpenetrate, squeezing the bulk medium molecules
`out of the inter-particulate space as illustrated in Fig. 3. This
`interpenetration is thermodynamically disfavored when a good
`solvent is used as the bulk medium to stabilize the tail [36].
`Accordingly, provided that the stabilizers can be absorbed onto the
`particle surface through the anchor segment, strong enthalpic
`interaction (good solvation) between the solvent and the stabilizing
`segment of the stabilizer is the key factor to achieve steric
`stabilization and prevent particles from agglomeration in the medium
`[36,37]. In addition to solvation, the stabilizing moiety needs to be
`sufficiently long and dense to maintain a steric barrier that is capable
`of minimizing particle–particle interaction to a level that the VDW
`attractive forces are less than the repulsive steric forces [43–45].
`The main drawback associated with the steric stabilization is the
`constant need to tailor the anchoring tail according to the particular
`drug of interest. Due to the lack of fundamental understanding of
`interaction between the stabilizers and dispersed nanoparticles,
`current surfactant screening approaches to achieve a successful steric
`stabilization are mostly empirical and could be very burdensome [45–
`49]. In addition, the solvation of the stabilizing segment is susceptible
`to variation in temperature. Stabilizer concentration could also play a
`role in causing suspension instability by affecting the absorption
`affinity of non-ionic stabilizers to drug particles surface. Deng et al.
`
`Fig. 1. Sedimentation in (a) deflocculated suspension; (b) flocculated suspension; and
`(c) open flocs-based suspension.
`
`and plate shaped itraconazole nanoparticles with aspect ratios
`between 5 and 10 [40].
`Although sedimentation is one of the key issues for colloidal
`suspension, the reported studies examining sedimentation issues in
`aqueous-based nanosuspensions are very scarce. This could be due to
`(i) surfactants are generally used in most of the nanosuspensions to
`inhibit particle agglomeration in the medium, which alleviates the
`sedimentation issues and (ii) the small nano-sized particles signifi-
`cantly reduce the sedimentation rate.
`In addition, many of the
`aqueous nanosuspensions are transformed to dry solid form by spray
`drying or freeze drying to circumvent the long-term sedimentation
`issue. Unfortunately, this solidification process cannot be applied to
`non-aqueous nanosuspensions where sedimentation/creaming is
`commonly present. An example to illustrate this is metered dose
`inhaler (MDI) formulations where the nanoparticles are suspended in
`HFA propellants. Sedimentation or creaming is a key aspect affecting
`stability of these formulations. Particle engineering to optimize
`particle surface properties and morphology, e.g. hollow porous
`particles [42], and introduction of surfactant(s) is generally employed
`to alleviate the issue.
`
`2.2.2. Agglomeration
`The large surface area of nanoparticles creates high total surface
`energy, which is thermodynamically unfavorable. Accordingly, the
`particles tend to agglomerate to minimize the surface energy.
`Agglomeration can cause a variety of issues for nanosuspensions
`including rapid settling/creaming, crystal growth and inconsistent
`dosing. The most common strategy to tackle this issue is to introduce
`stabilizers to the formulation. In addition to safety and regulation
`
`

`

`L. Wu et al. / Advanced Drug Delivery Reviews 63 (2011) 456–469
`
`459
`
`Fig. 2. Illustration of classical DLVO theory. Attractive forces are dominant at very small and large distances, leading to primary and secondary minimum, while repulsive forces are
`prevailing at intermediate distances and create net repulsion between the dispersed particles, thus preventing particle agglomeration.
`
`[50] used Pluronic® F127 to stabilize paclitaxel nanosuspensions. It
`was reported that stabilizers had high affinity to nanocrystals surface
`at concentrations below critical micelle concentration (CMC), and
`increasing concentrations above CMC caused loss of F127 affinity to
`the nanocrystals and thus unstable formulation. This was because
`F127 monomers on the nanocrystals surface started to aggregate with
`each other to form micelles as the concentration was increased to the
`CMC level, leading to a lower affinity to the drug crystals. Temperature
`was also shown to affect the stabilizer affinity to drug crystals. This is
`expected since the CMC level is dependent on temperature.
`It is apparent that combination of the two stabilization mechanisms
`can be very beneficial in achieving a stable colloidal dispersion. In
`addition, the combination of a non-ionic stabilizer with an ionic
`stabilizer reduces the self repulsion between the ionic surfactant
`molecules, leading to closer packing of the stabilizer molecules [10,51].
`
`Besides the steric and electrostatic stabilization mechanisms, some
`other stabilization mechanisms have also been reported. Makhlof et
`al. produced indomethacin (IMC) nanocrystals using the emulsion
`solvent diffusion technique [52]. The nanoparticles were stabilized
`using various cyclodextrins (CyDs) without adding any surfactants.
`The stabilizing effect was attributed to the formation of a CyD network
`in the aqueous medium via intermolecular interaction of CyD
`molecules. The network-like structure was believed to prevent
`aggregation and crystal growth of
`IMC nanoparticles initially
`produced from the solvent diffusion process. Similar stabilization
`mechanism was also observed in another study where budesonide
`microsuspension was stabilized with hydroxypropyl-beta-cyclodex-
`trin in HFA medium [53]. Another approach to enhance suspension
`stability that has increasingly been utilized is engineering of particle
`morphology. One breakthrough in this area was the porous particle
`
`Fig. 3. Steric stabilization mechanisms according to Gibbs free energy: ΔG = ΔH−TΔS. A positive ΔG indicates stable suspension while negative ΔG induces particle agglomeration. If
`the medium is a good solvent for the stabilizing moiety, the adsorbed stabilizing layers on the dispersed particles cannot interpenetrate each other when the particles collide. This
`reduces the number of configurations available to the adsorbed stabilizing tails, resulting in a negative entropy change and positive ΔG. On the other hand, if the dispersion medium
`is a poor solvent, the adsorbed layers on the particles may interpenetrate thermodynamically and induces particles agglomeration.
`
`

`

`460
`
`L. Wu et al. / Advanced Drug Delivery Reviews 63 (2011) 456–469
`
`concept that was first introduced by Edwards et al. [54]. The porous
`particles include hollow porous particle [42] and porous nanoparticle-
`aggregate particles (PNAPs) [14]. Unfortunately, most of the work has
`been focused on microsuspension or polymeric colloidal formulations
`and has not been applied to pure drug nanoparticles.
`Table 1 summarizes a few published studies on pharmaceutical
`nanosuspensions. Due to the vast amount of literature work on the
`pharmaceutical nanosuspensions, this review will focus only on the
`studies that provide a more profound enlightenment on the stabilizer
`selection for nanosuspensions. The summary table shows that most of
`nanosuspensions were generated in aqueous medium, with only a
`limited number of nanosuspensions made in non-aqueous environ-
`ment. The commonly used ionic stabilizers in aqueous medium include
`sodium dodecyl sulfate (SDS), sodium lauryl sulfate (SLS), lecithin and
`docusate sodium. The non-ionic surfactants used in aqueous medium
`are usually selected from Pluronic® surfactants, Tween 80, polyeth-
`ylene glycol (PEG), polyvinyl alcohol (PVA) polyvinylpyrrolidone
`(PVP) and cellulose polymers such as hydroxypropyl cellulose (HPC)
`and hydroxypropyl methylcellulose (HPMC).
`The stabilizers are not only used to provide short- and long-term
`storage stability for nanosuspensions, but also to achieve successful
`formation and stabilization of nanocrystals during particle produc-
`tion. Lee et al. designed and synthesized various amino acid
`copolymers containing lysine as the hydrophilic segments with
`alanine, phenylalanine or leucine as hydrophobic moieties [49]. Wet
`comminution was used to produce naproxen nanosuspensions in
`presence of HPC and amino acid copolymers. Lysine copolymer with
`alanine was unable to produce submicron particles while the other
`copolymers with phenylalanine and leucine were capable of forming
`the nanoparticles. The size of nanocrystals was proven to be constant
`over 1 month storage and the crystallinity was also shown to be
`preserved after the wet comminution process. Furthermore, hydro-
`phobicity of the copolymers was identified as the key factor in
`achieving the stable nanosuspensions, attributed to strong polymer
`adsorption onto the hydrophobic drug surfaces. Although this work
`did not provide an in-depth discussion on how the copolymers
`interacted with the drug nanoparticles, it illustrated the importance of
`careful selection of the anchor group (that is attached to the drug
`surface) in facilitating the production of a stable nanosuspesion. In the
`subsequent study [45], they attempted to understand the nature of
`interactions between polymeric stabilizers and drugs with different
`surface energies. Nanocrystals of seven model drugs with PVP K30 and
`HPC as stabilizers were generated using wet comminution. It was
`expected that a close match of surface energy between the stabilizers
`and drug crystals would promote the absorption of stabilizers onto
`drug particles, and thus help in reducing the particle size during the
`wet comminution process. Although surface energy did not seem to
`correlate well with particle size for HPC stabilized system, some trend
`was observed for PVP stabilized suspension with only one exception.
`A further study with seven stabilizers (non-ionic stabilizers: HPC,
`PVP K30, Pluronic® F127 & F68, PEG and ionic stabilizers: SDS and
`benzethoinum chloride) and eleven model drugs was conducted by
`the same group in order to provide more understanding on the
`stabilization mechanism [48]. Again, the general trend between
`surface energy and particle size reduction was not observed in this
`work. PEG was unsuccessful in reducing the particle size of most drug
`candidates while the other non-ionic stabilizers proved to be effective
`in reducing the size of five drug candidates that had similar surface
`energies to the stabilizers. F68 was shown to be the most effective
`stabilizer (successfully stabilizing nine drug candidates), which could
`be due to its strong chain adsorption onto the drug crystals through
`the hydrophobic polypropylene glycol (PPG) units. F127 was found to
`be less efficient than F68 likely because the short processing time led
`to inefficient physical adsorption of higher molecular weight F127 to
`the drug surface. This study demonstrated that a combination of ionic
`and non-ionic stabilizers is not always beneficial to enhance
`
`stabilization, A few combinations of SDS or benzethoinum chloride
`with various non-ionic stabilizers resulted in positive stability effects
`while the others did not. The effects of physicochemical properties of
`the drugs on the stabilization were also explored in this study. In
`general, drugs with lower aqueous solubility, higher molecular weight
`and higher melting point were shown to have higher chance for
`successful nanosuspension formation.
`Van Eerdenbrugh et al. conducted an expanded study using 13
`stabilizers at 3 different concentrations to stabilize 9 drug compounds
`[47]. The particles were generated using the wet milling technique.
`The success rate in producing nanosuspensions using polysaccharide
`based stabilizers [HPMC, methylcellulose (MC), hydroxyethylcellulose
`(HEC), HPC, carboxymethylcellulose sodium (NaCMC), alginic acid
`sodium (NaAlg)] was limited by the high viscosity of these polymeric
`stabilizer solutions. Increasing concentration of these stabilizers did
`not appear to be helpful. In contrast, the other stabilizers [PVP K30,
`PVP K90, PVA, Pluronic® F68, polyvinyl alcohol–polyethylene glycol
`graft copolymer (K-IR), Tween 80 and D-α-tocopherol polyethylene
`glycol 1000 succinate (TPGS)] did not encounter the viscosity issue.
`PVA was ineffective in producing the nanosuspension and the success
`probability of PVP K30, PVP K90, F68 and K-IR is highly dependent on
`their concentration. Higher concentrations (25 wt.% and 100 wt.%)
`increased the stabilizing efficacy significantly. Tween 80 and
`TGPS were proven to be the most effective stabilizers. Addition of
`TGPS (at concentrations N25 wt.%) allowed nanosuspension forma-
`tion for all tested drug compounds. No correlation was observed
`between drug physicochemical properties (molecular weight,
`melting point, log p, solubility and density) and nanosuspension
`formation success rate. It was demonstrated that surface hydropho-
`bicity of the drug candidates was the driving force for nanoparticles
`agglomeration, thus lowering the success rate of nanosuspension
`production.
`Mishra et al. explored nanosuspension stability issues during both
`production and storage [29]. Hesperetin nanosuspensions were
`produced using HPH with Pluronic® F68, alkyl polyglycoside
`(Plantacare 2000) and inulin lauryl carbamate (Inutec SP1), or
`Tween 80 as stabilizers. It was demonstrated that all stabilizers
`were suitable for successful production of hesperetin nanosuspen-
`sions. The size of nanocrystals was dependent on power density
`applied in the homogenization process and the hardness of the
`crystals. The effect of stabilizers on the particle size was negligible.
`Short-term stability over a period of 30 days was examined in order to
`evaluate the stabilizer efficiency. ZP was measured as a key parameter
`to predict the stability. In distilled water, the ZP values of all the
`nanosuspensions fell between −30 and −50 mV and the values
`dropped significantly in the original dispersion medium. This can be
`explained by the fact that adsorbed layers of large molecules shifted
`the shear plane to a longer distance from the particle surface, thus
`reducing the measured value of zeta potential (Fig. 4). However, the
`low ZP value does not point to an unstable suspension in this case,
`which could be due to the additional presence of steric stabilization
`mechanism. Both Inutec and Plantacare stabilized nanosuspensions
`also showed significant reduction of ZP measured from water to
`dispersion medium, indicating a thick absorbed steric layer and good
`stability. F68 exhibited only slight decrease in ZP,
`indicating a
`relatively thin stabilization layer. The ZP value of Tween 80 was
`only −13 mv in the dispersion medium, pointing to a potentially
`problematic stabilization. The study demonstrated that zeta potential
`measurement is a good predictor for storage stability. Nanosuspen-
`sions stabilized by Inutec and Plantacare were stable at all storage
`conditions (4, 25 and 40 °C) up to 30 days while F68 stabilized
`nanosuspensions were shown to be less stable. The Tween 80
`formulation stability was the poorest. Pardeike et al. [30] conducted
`a similar study using phospholipase A2 inhibitor PX-18 nanosuspen-
`sions produced by HPH with Tween 80 as stabilizer. In this work, ZP of
`the homogenized nanosuspensions was dropped from −50 mV to
`
`

`

`L. Wu et al. / Advanced Drug Delivery Reviews 63 (2011) 456–469
`
`461
`
`Table 1
`Literature summary of pharmaceutical nanosuspensions.
`
`Nanoparticles compound
`
`Manufacturing
`technique
`
`Delivery
`route
`
`Dispersion
`medium
`
`Stabilizers
`
`Reference
`
`Oridonin
`Oridonin
`Budesonide
`Buparvaquone
`Buparvaquone
`Diclofenac acid
`Azothromycin
`Rutin
`Rutin
`Tarazepide
`Omeprazole
`Amphotericin B
`Nimodipine
`Albendazole
`RMKP 22
`Hesperetin
`
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`
`HPH
`
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`HPH
`Wet milling
`
`Wet milling
`Wet milling
`Wet milling
`Wet milling
`Wet milling
`Wet milling
`Wet milling
`
`Hydrocortisone, prednisolone
`and dexamethasone
`Ascorbyl palmitate
`RMKK99
`Nifedipine
`Undisclosed
`Hydroxycamptothecin
`Asulacrine
`RMKP 22
`RMKP 22
`PX-18
`PX-18
`Silybin
`Tarazepide
`Omeprazole, albendazole
`and danazol
`Fluticasone, budesonide
`Naproxen
`Loviride
`Nine different compounds
`Zinc Insulin
`Ethyl Diatrizoate
`Cinnarizine, itraconazole
`and phenylbutazone
`Wet milling
`Nine different compounds
`Beclomethasone dipropionate Wet milling
`Rilpivirine
`Wet milling
`Undisclosed
`Wet milling
`Piposulfan, etoposide,
`Wet milling
`camptothecin, paclitaxel
`Naproxen
`Seven different compounds
`Eleven different compounds
`
`Wet comminution
`Wet comminution
`Wet comminution
`
`Water
`NA
`Water
`IV
`Inhalation Water
`Inhalation Water
`Oral
`Water
`Oral
`Water
`NA
`Water
`Oral
`Water
`Oral
`Water
`NA
`Water
`IV
`Water
`Oral
`Water
`IV
`Water
`Oral
`Water
`NA
`Water
`Dermal
`Water
`
`Opthalmic Water
`
`NA
`NA
`NA
`Oral
`NA
`IV
`NA
`NA
`NA
`NA
`Oral, IV
`IV
`Oral
`
`Water
`Water
`Water
`Water
`Water
`Water
`Water
`Water
`Water
`Water
`Water
`Water
`Water
`
`Inhalation Water
`NA
`Water
`NA
`Water
`NA
`Wate