1.
`
`2.
`
`3.
`
`Introduction
`
`Physical forms of the
`formulated drugs
`
`Formulations for compound
`selection
`
`4. Hidden poorly
`water-soluble compounds: the
`salt issue
`
`5.
`
`6.
`
`7.
`
`8.
`
`Intellectual property
`
`Status of delivery technologies
`for poorly
`water-soluble compounds
`
`Regulatory aspects
`
`Expert opinion
`
`Review
`
`Drug delivery strategies for poorly
`water-soluble drugs: the
`industrial perspective
`
`†
`Peter van Hoogevest, Xiangli Liu & Alfred Fahr
`†
`Institute for Pharmacy, Department of Pharmaceutical Technology, Friedrich-Schiller-Universita¨t
`Jena, Jena, Germany
`
`Introduction: For poorly soluble compounds, a good bioavailability is typically
`needed to assess the therapeutic index and the suitability of the compound for
`technical development. In industry, the selection of the delivery technology is
`not only driven by technical targets, but also by constraints, such as production
`costs, time required for development and the intellectual property situation.
`Areas covered: This review covers current developments in parenteral and
`oral delivery technologies and products for poorly water-soluble compounds,
`such as nano-suspensions, solid dispersions and liposomes. In addition, the use
`of biorelevant dissolution media to assess dissolution and solubility properties
`is described. Suggestions are also included to systematically address develop-
`ment hurdles typical of poorly water-soluble compounds intended for
`parenteral or oral administration.
`Expert opinion: A holistic assessment is recommended to select the appropri-
`ate delivery technology by taking into account technical as well as intellectual
`property considerations. Therefore, first and foremost, a comprehensive
`physico-chemical characterization of poorly water-soluble compounds can
`provide the key for a successful selection and development outcome. In this
`context, the identified physical form of the compound in the formulation is
`used as a guide for a risk--benefit assessment of the selected oral delivery tech-
`nology. The potential of nano-suspensions for intravenous administration is
`unclear.
`In the case of oral administration, nano-suspensions are mainly
`used to improve the oral absorption characteristics of micronized formula-
`tions. The development of an in situ instantaneous solubilization method,
`based on stable, standardized liposomes with low toxicity, opens new
`avenues to solubilize poorly water-soluble compounds.
`
`Keywords: biorelevant solubilization, drug delivery, liposomes solubilization, nano-suspension,
`poorly water-soluble drug, poorly water-soluble drug dissolution, solubilization
`
`Expert Opin. Drug Deliv. (2011) 8(11):1481-1500
`
`1. Introduction
`
`Poorly water-soluble drugs continue to present challenges for oral and parenteral
`delivery. After oral administration, the degree of absorption may be highly variable
`and unreliable, thus, requiring suitable formulations to improve bioavailability. To
`allow parenteral (intravenous; i.v.) administration, the desired dose of the drug
`should be delivered in a solution type of formulation which prevents precipitation
`at the injection site and further dilution in the bloodstream. In sharp contrast to
`water-soluble drugs, where the formulations are used to facilitate processing and
`accurate dosing, formulations for many poorly water-soluble compounds should
`also increase bioavailability. For such insoluble compounds, a successful formulation
`is,
`therefore, absolutely essential
`to develop an efficacious drug product.
`
`10.1517/17425247.2011.614228 © 2011 Informa UK, Ltd. ISSN 1742-5247
`All rights reserved: reproduction in whole or in part not permitted
`
`1481
`
`

`

`Drug delivery strategies for poorly water-soluble drugs: the industrial perspective
`
`Article highlights.
`
`. The physical form of a poorly water-soluble compound
`can guide the selection of the formulation technology
`for the compound.
`. Characterization of the poorly water-soluble compound
`should precede formulation screens.
`. Drugs and formulations intended for oral administration
`are characterized by reliable solubility and dissolution
`testing in reproducible biorelevant dissolution media
`prepared from standardized powders.
`. Liposomes can be considered as a solubilization
`technique for poorly water-soluble compounds due to
`the availability of standardized in situ
`solubilization methods.
`. Nano-suspensions are well suited to develop line
`extensions and eliminating food effects. For early
`research work, dependent on the availability of
`nano-milling equipment and severity of and
`lack of oral absorption, the suitability of
`nano-suspensions may be explored in comparison
`with micronized formulations.
`. Crystalline nano-suspensions are not generally suitable
`for intravenous administration given their slow release
`and dissolution properties. However, they can be
`considered as depot formulations for intramuscular or
`subcutaneous administrations.
`. Amorphous nano-suspensions using albumin as a
`matrix-excipient are suitable for intravenous delivery as
`proven by the product Abraxane.
`. The use of cyclodextrins suitable for intravenous
`administration and nano-suspension technologies may
`be protected by certain patents. The commercial terms
`associated with the intellectual property may influence
`the selection of the commercial formulation for poorly
`water-soluble compounds.
`. Due to the increasing number of poorly water-soluble
`drug candidates, formulations may require the use of
`volatile organic solvents (solid dispersions, liposomes and
`amorphous nano-suspensions).
`. Regulatory authorities are developing new guidance to
`deal with special delivery technologies related to
`nanotechnology. This guidance should be considered
`when compiling CMC documentation.
`
`This box summarizes key points contained in the article.
`
`Underestimating the importance of formulations for poorly
`water-soluble drugs may result in the selection of the wrong
`lead candidates, prolong development timelines and signifi-
`cantly increase cost of development. To enable a proper selec-
`tion of compound and corresponding formulation,
`the
`compounds
`require special attention regarding physico-
`chemical characterization. A close interaction between chemi-
`cal development (responsible for the supply of the drug
`substance in a chemically and physically well-defined form)
`and pharmaceutical development (responsible for incorporat-
`ing the drug in the dosage form to make the drug product) is
`crucial. It ensures that there is a good technological match
`between drug and dosage form early on in development.
`
`Although poorly water-soluble compounds possess a repu-
`tation of being cumbersome development candidates, drug
`lipophilicity is often necessary for efficient interaction with
`the target
`receptor. Furthermore,
`lipophilicity is also a
`requirement to pass the lipidic domain of natural (bilayer-
`phospholipid) membranes. Also, seemingly highly water-
`soluble drugs which are delivered in salt form can pass the
`membrane in the uncharged lipophilic base or acid form
`which is in equilibrium with the salt form in the water phase.
`The lack of water solubility does not necessarily prevent
`poorly water-soluble compounds from distributing through
`the body after oral absorption or parenteral administration.
`Poorly water soluble, lipophilic compounds use a ‘lipid high-
`way’ which involves reversible binding to circulating lipid car-
`riers such as albumin and lipoproteins and possibly blood
`cells, thereby transporting the compound to the target recep-
`tor. The exchange/transfer of the compound between the var-
`ious lipid domains,
`including the lipidic domain of the
`formulation, may occur through diffusion and/or collision
`mechanisms [1].
`There are multiple delivery options for oral administration
`of poorly soluble drugs, whereas for parenteral administration
`there are a comparatively limited number of
`suitable
`approaches. In addition, related to the wide interest on nano-
`technology, there are an increasing number of research ideas
`which may also be suitable. Simultaneously,
`regulatory
`authorities such as the FDA and EMA (European Medicine
`Agency) are becoming increasingly aware of the complications
`and quality requirements related with drug delivery tech-
`nologies for poorly water-soluble compounds [2-7]. Also, the
`intellectual property (IP) situation and patentability of formu-
`lation play a paramount role for the industrially oriented sci-
`entist in selecting a particular delivery technology. This factor
`is often ignored by most if not all scientific papers and is all
`the more remarkable considering that patents are the source
`of scientific innovation. The industrial pharmaceutical tech-
`nologist is often confronted with an increasingly complex
`jungle of options to select the ‘best’ formulation/delivery
`option. Both scientific and commercial perspectives have to
`be considered to satisfy increasingly challenging quality and
`development requirements.
`From a financial perspective, unlike classical solid dosage
`form manufacturing processes, for example, granulation and
`tabletting, bioavailability enhancing formulation technologies
`for poorly soluble compounds are usually more expensive.
`They require significant additional capital expenditure and
`resource investments. This may be more acceptable for line
`extensions which is already generating revenues but can be
`risky for new chemical entities (NCEs) which are yet to
`be commercialized.
`This review article is an update of a previously published
`review article [8]. It summarizes the latest advancements in for-
`mulation strategies for oral as well as parenteral administra-
`tion of poorly water-soluble compounds and especially
`taking industrial aspects into consideration.
`
`1482
`
`Expert Opin. Drug Deliv. (2011) 8(11)
`
`

`

`2. Physical forms of the formulated drugs
`
`A detailed physico-chemical characterization of a potential
`therapeutic compound should start before formulation experi-
`ments or investing financially into its future. This seems obvi-
`ous and is indeed common practice in big pharmaceutical
`companies. However, its importance is frequently underesti-
`mated in small pharmaceutical companies, which perhaps
`are unaware of the development challenges poor solubility
`can present or are financially constrained. The recommended
`minimal characterization of poorly water-soluble drug sub-
`stance, relevant for oral formulation development, analogous
`to [9] is provided in Table 1.
`If i.v. administration is the target, some characteristics do
`not need to be assessed for pharmaceutical development in
`as much detail, because the drug is in the solubilized form.
`The results of the characterization serve as a basis to decide
`rationally whether the drug is suitable at all for further formu-
`lation development, estimate development efforts and which
`delivery approach(es) could be explored to maximize bioavail-
`ability. Undesirable properties are, for example, lack of stabil-
`ity in the solid state, multiple and unpredictable polymorphs,
`extreme hygroscopicity and light sensitivity, extremely low
`solubility and (if oral administration is envisaged) instability
`in (simulated) physiological media such as simulated gastric
`fluid (SGF), fasted state simulated intestinal fluid (FaSSIF)
`and fed state simulated intestinal fluid (FeSSIF).
`Drug purity is also a key. As a cautionary note, to identify a
`development path for future formulation options, experi-
`ments performed with 95% or less pure drug substances
`should be considered with extreme caution. This is because
`the impurities may influence the therapeutic/toxicity profile
`and influence the physical form of the drug substance and
`related solubility characteristics. At this point, a complete false
`decision base may be created.
`Based on the previously performed compound profiling, it
`is now of importance to select possible formulation strategies
`to be able to examine the in vivo efficacy and biopharmaceu-
`tical behavior (absorption, distribution, metabolism, elimina-
`tion and toxicity; ADMET) of the compound. When the
`therapeutic index of a series of poorly water-soluble com-
`pounds has to be compared, in principle, optimal formula-
`tions for every compound are needed. In the literature
`various, somewhat complex, decision trees have been pro-
`posed (see for a typical example [10]). In order to enable a
`more systematic approach to assess which formulation strat-
`egy may be best for a drug, a correlation between the physical
`form of the drug substance in the formulation and possible
`parenteral and oral formulation technologies is proposed for
`consideration (Figure 1).
`The drug may be in solubilized, crystalline or amorphous
`form in the formulation. In this context, ‘solubilized’ encom-
`passes mono-molecular dispersion/solutions of the poorly
`water-soluble compounds in the formulation. The presence
`of mixtures of the drug in a different physical form may result
`
`van Hoogevest, Liu & Fahr
`
`in physical instability (e.g., crystalline suspension in presence
`of relatively high concentration of solubilized drug may give
`rise to Oswald ripening; presence of crystalline seeds in amor-
`phous material may trigger crystallization of the amorphous
`fraction). The physical form of the drug can be used as a trans-
`parent guide to assess and understand the pros and cons of the
`various formulation options/technologies. It encourages an
`intuitive categorization of the technical risks associated with
`the physical form of the selected formulation approach.
`
`2.1 Solubilized form
`Technologies/formulations comprising the solubilized form
`of the drug substance may be suitable for oral drugs where
`oral absorption is not limited by the passage (permeability)
`through the intestinal epithelial membranes. For i.v. adminis-
`tration the solubilized form is clearly preferred. The solubi-
`lized form has the inherent advantage that the drugs are
`already in solution and no dissolution step is necessary to
`become bioavailable. Disadvantage of the solubilized form
`may be the lack of long-term (chemical) stability and the
`risk for drug precipitation on dilution in the biological milieu.
`The risk of precipitation is, however, dependent on how the
`drug is solubilized. In case of, for example, using water misci-
`ble solvents, surfactants or pH adjustment (or using the salt
`form of the drug) to dissolve drugs, the chance that dilution
`and restoring physiological pH will result in precipitation
`will be higher compared to solubilization technologies in
`which the drug is dissolved in water immiscible dispersed oil
`phases (e.g., oil-in-water emulsions and liposomes). The
`degree of dissolution in the formulation should be sufficiently
`high to enable the co-administration of low (non-toxic and
`physiologically acceptable) amounts of the excipients.
`There are two main solubilization types, water miscible and
`water immiscible. Water miscibles are exemplified by organic
`solvent/water mixtures, aqueous pH adjustment, aqueous
`cyclodextrin, surfactant and mixed micellar solutions. Water
`immiscible solubilization types are exemplified by oily
`solutions, oil-in-water emulsion (macro emulsion, self-emulsi-
`fying systems) and liposomes. In the water miscible type, the
`solubilizing capacity for the poorly water-soluble compound
`in the aqueous medium decreases because of dilution of the
`solubilizing component (solvent or surfactant) and as a result
`the drug may precipitate in the excess of water phase
`(i.e., blood circulation or gastrointestinal liquid). With the
`water immiscible type, where the drug is associated and
`solubilized in the oily component of the formulation, the
`formulation does not lose its solubilizing potential on dilu-
`tion with water as long as the oil/lipophilic component
`remains intact.
`
`2.2 Amorphous form
`The amorphous form in the formulation (e.g., solid dispersion)
`may be considered when the drug cannot be suitably solubi-
`lized. On hydration of a formulation containing the amorphous
`form, supersaturation may occur where concentrations are
`
`Expert Opin. Drug Deliv. (2011) 8(11)
`
`1483
`
`

`

`Drug delivery strategies for poorly water-soluble drugs: the industrial perspective
`
`Table 1. Physico-chemical characterization of poorly
`water-soluble drug before starting oral formulation
`screen.
`
`Characteristic
`
`Method
`
`Basic information
`Purity
`Appearance and crystal
`habit and size*
`UV-Vis spectrum
`Crystal form
`Melting point
`pKa
`Log P or D
`Solubility
`Solvents
`(water miscible/immiscible)
`Aqueous media pH 1 -- 10
`SGF, FaSSIF, FeSSIF
`Accelerated stability
`pH (solvent water mixture)
`Light (solvent water mixture)
`Heat (solvent water mixture
`and solid)
`Mechanical stress*
`Crystal form in dispersion*
`
`Hygroscopicity
`
`HPLC
`Visual inspection and
`microscopy
`Spectrophotometry
`XRD or Raman spectroscopy
`DSC
`Titrimetric methods
`HPLC
`
`HPLC or UV-Vis
`
`HPLC or UV-Vis
`HPLC or UV-Vis
`
`HPLC
`HPLC
`HPLC
`
`XRD or Raman spectroscopy
`XRD or Raman spectroscopy
`or microscopy
`Visual inspection/Karl
`Fischer/Raman spectroscopy
`
`*Not needed in case of intended intravenous administration.
`DSC: Differential scanning calorimetry; FaSSIF: Fasted state simulated intestinal
`fluid; FeSSIF: Fed state simulated intestinal fluid; SGF: Simulated gastric fluid;
`UV-Vis: UV-visible spectrometry; XRD: X-ray diffraction.
`
`higher than achievable when formulations containing the crys-
`talline drug are dissolved. The supersaturated solution may
`give rise to an increased oral absorption. The use of the amor-
`phous form is clearly a double edged sword: on one hand, the
`possibility to increase oral absorption and on the other, a pos-
`sible lack of chemical and physical stability and risk of crystal-
`lization. The development of such dosage forms, therefore,
`requires extensive real time stability testing accompanied by
`intensive quality control (QC) to monitor possible onset of
`crystallization of the drug. In addition, the production of
`such dosage forms either require the use of melt extrusion or
`hot melt technologies or spray drying from solvents (examples
`of solid dispersion products are given below). Melt technolo-
`gies may be limited when the drug is very heat sensitive. Spray
`drying can be cumbersome due to the use of solvents and fur-
`ther development into a solid dosage form. An oral dosage
`form comprising the amorphous form of the drug may be
`very useful for oral toxicity testing because supersaturated solu-
`tions may give a higher oral absorption to maximize exposure
`of the animal to the drug. However, also in this case robust sta-
`bility data are needed to guarantee that the drug during the
`testing period will stay in the preferred amorphous state.
`Immobilization of the poorly water-soluble compound in
`highly porous inorganic carriers has an interesting delivery
`
`potential which may combine the amorphous form of the
`drug and increase of surface area to further increase the
`dissolution rate [11,12].
`The amorphous form of the poorly water-soluble drug pac-
`Ò
`is used for i.v. administra-
`litaxel in the product Abraxane
`tion [13]. This is the only commercially available example of
`i.v. administration of a solid (amorphous) form of a poorly
`water-soluble drug.
`
`2.3 Crystalline form
`The most robust form of the drug substance, from a chemical
`stability perspective, which can be used to develop a dosage
`form for a poorly water-soluble drug is the crystalline form.
`To be orally bioavailable the solid crystalline form first has
`to dissolve. The dissolution rate of the drug substance (for
`pH insensitive compounds) depends on: i) the crystal form
`being used and ii) the particle size. The selection of the crystal
`form is of paramount
`importance. Omitting polymorph
`screens entirely at an early development stage on the drug sub-
`stances may cause serious development problems during later
`development stages. To increase the dissolution rate after oral
`administration, the smallest possible particle sizes of dispersed
`drug, in the form of nano-suspensions, have been considered
`for at least 20 years. These efforts indeed resulted in a few
`products (see below). So far, only one crystalline nano-
`suspensions for subcutaneous and intramuscular (i.m.) has
`been successfully commercialized but i.v. formulations using
`this approach have yet to be commercialized.
`
`3. Formulations for compound selection
`
`3.1 Parenteral
`For an initial comparison of the in vivo profile of several drug
`candidates against each other in a family of compounds or an
`individual candidate, the parenteral (i.v.) route is preferred.
`This route eliminates uncertainties regarding assessment of
`degree of efficacy, related to an (unknown) fraction of
`absorbed oral dose. However, in practice, the selection of an
`adequate i.v. formulation is sometimes an impassable hurdle.
`Formulations containing the solubilized form of the com-
`pound (Figure 1) which on dilution in aqueous medium (sim-
`ulating blood) do not
`show precipitation are preferred.
`Uncontrolled precipitation of drugs may give rise to a signifi-
`cant reduction of the absolute bioavailability and consequently
`an overestimation of the degree of oral bioavailability. Clearly,
`comparison of the efficacy and pharmacokinetics (PK) of sev-
`eral drug candidates with different precipitation tendencies
`may give rise to a completely false selection procedure.
`In general, for any of the formulation options, excipients
`which are well established and accepted by regulatory author-
`ities should be preferred. Preclinical research with cocktails of
`exotic excipients may initially result in seemingly adequate
`formulations, but sooner or later the formulation has to be
`reformulated to make it more acceptable for clinical research.
`Delay in switching to more reliable formulations can be more
`
`1484
`
`Expert Opin. Drug Deliv. (2011) 8(11)
`
`

`

`van Hoogevest, Liu & Fahr
`
`Intravenous
`
`Solubilized
`
` pH adjustment (in-situ salt)
` Co-solvent
` Complexed with cyclodextrins
` Surfactants
` Mixed micelles
` Oil-in-water emulsion
` Liposomes
`
`Oral
`
`Solubilized
`
` pH adjustment (in-situ salt)
` Co-solvent
` Complexed with cyclodextrins
` Surfactants
` Oil
` Mixed micelles
`Oil-in-water emulsion: macro-
`emulsion and micro-emulsion
` Liposomes
` Solid lipid particles
`
`Amorphous
`
`Nanosuspension/
`hydrosol*
`
`Crystalline
`
`Nanosuspension/
`hydrosol‡
`
`Amorphous
`
`Nanosuspension/
`hydrosol*
`
`Crystalline
`
`Nanosuspension/
`Hydrosol‡
`Microsuspension
`
`Figure 1. Categorization of formulation options for intravenous and oral poorly water-soluble drugs based on the physical
`form of the drug in the formulation.
`*Prepared using a solvent dilution method.
`z
`Prepared using milling of crystalline drug.
`
`costly in coping with the consequences (e.g., repeating tox
`testing, stability testing and even clinical trials). The choice
`of excipients in preclinical research is also technologically
`(e.g., compatibility) and biologically influenced (e.g., the
`target animal and disease model and the non-observed adverse
`
`the formulation/
`further development of
`level). If
`effect
`product
`is envisioned, commercial
`factors
`such as cost-
`of-goods, manufacturing costs and availability, guarantee of
`world-wide supply, the patent status and associated commercial
`terms and conditions need to be considered as well.
`
`Expert Opin. Drug Deliv. (2011) 8(11)
`
`1485
`
`

`

`Drug delivery strategies for poorly water-soluble drugs: the industrial perspective
`
`The pros and cons of the i.v. formulation options (Figure 1)
`are now critically reviewed. Formulation development costs to
`administer intravenously poorly water-soluble compounds
`may be reduced in preclinical development by using the
`same DMSO solution of the drug as used for drug discovery
`in in vitro screens. This approach is, however, not recom-
`mended, because DMSO is rather toxic and the dilution of
`the DMSO may result in uncontrollable drug precipitation
`in the blood circulation. PEG solutions, although less toxic,
`also do not prevent precipitation. Attempts to make the use
`of PEG solutions acceptable by injecting the PEG solution
`very slowly (to reduce the risk of precipitation) may look
`problem solving but carry the risk that the resulting PK data
`may depend on the injection rate and concomitant precipita-
`tion tendency. Also, the use of N,N dimethylacetamide or
`N-methylpyrrolidone, justified by the existence of two FDA
`approved products which contain these rather exotic solvents
`Ò
`Ò
`at low concentrations (Busulfex
`and Eligard
`), is not recom-
`mended. Ethanol and propylene glycol may be considered but
`can only be used at 10 -- 20% in aqueous vehicles. However,
`with high water content, the ability to keep very poorly solu-
`ble compounds in solution may be limited.
`If compounds only dissolve at non-physiological pH by
`adjustment to acidic or alkaline pH, there will be a risk that
`the compounds may precipitate after mixing with blood at
`the injection site and/or in the circulation. Extreme pH can
`cause phlebitis and in vivo precipitation may cause embo-
`lisms. This approach looks simple but may ignore risks that
`at a later phase clinical testing, such as unacceptable pain
`and tolerability at the injection site experienced by patients,
`may put a brake for further development.
`Solubilization with surfactants, for example, Tween 20 or
`Ò
`EL, polyoxyl 35 cas-
`80 (polysorbate 20 and 80), Cremophor
`tor Oil (USP/NF) and Solutol HS 15 (PEG 660-hydroxy
`stearate), are also an option but they possess anaphylactic
`properties [14-23]. On dilution in aqueous media (infusion sol-
`utions or blood), the surfactant concentration may decrease
`below levels at which the drug dose can be solubilized. In
`addition, the general use of Solutol HS 15 in development
`may be hampered by the fact that there is no registered prod-
`uct with Solutol HS 15 on the market in the US. The FDA,
`therefore, considers Solutol HS 15 as a novel excipient.
`The use of complexing agents such as cyclodextrins can also
`be considered. Only two cyclodextrins are at present accept-
`able for parenteral use, hydroxypropyl-b-cyclodextrin and sul-
`fobutyl-b-cyclodextrin sodium salt. Both excipients are
`protected by some IP rights as elaborated below. It is very
`rare that research NCEs can be complexed with lower than
`10 molecules of cyclodextrin excess to drug. For these reasons,
`at especially high-dose testing during toxicity investigations,
`cyclodextrins may cause prohibitive side effects.
`Oil-in-water emulsions may be another option but this
`requires a very high solubility of the poorly water-soluble
`compounds in the oil phase to achieve adequate concentrations
`in 10 -- 20% oil emulsions. In addition, the oil emulsions may
`
`be destabilized by the drug load. In practice, most poorly
`water-soluble compounds also possess a rather low oil solubil-
`ity. As a result, this formulation technology is useful only in
`exceptional cases.
`Mixed micelles comprising bile salts and lecithin are being
`used to solubilize lipophilic vitamins for i.v. administration
`Ò
`MM pediatric (2.2%), Konakion (1.9%) and
`(Konakion
`Ò
`Cernevit
`, multivitamins for infusion). In spite of low toxic-
`ity [24,] they are not abundantly being used and hardly men-
`tioned in key reviews [25]. It should be noted that on dilution
`(in infusion bags or on i.v. administration), these formulations
`could undergo conversion from micelles to liposomes.
`Liposomes also have the potential to solubilize drugs. In
`contrast to micelles (e.g., polysorbate) and mixed micelles,
`they do not (immediately) convert into other physical struc-
`tures on dilution in aqueous media. The release of poorly
`water-soluble compounds occurs through diffusional or colli-
`sional transfer to other lipid components, such as lipoproteins,
`in the ‘lipid highway’ of blood [1,26]. In several review papers on
`the preclinical use of vehicles for poorly water-soluble com-
`pounds, liposomes are only considered as last option [25,27-32]
`despite the low toxicity of the used phospholipid excipients.
`This is probably because of the efforts to make tailored drug
`loaded liposomes at the preclinical level. Because of the avail-
`ability of ready-for-use liposomes with defined particle size
`and instantaneous loading procedures with poorly water-
`soluble drugs [33], liposomes are certainly useful options to
`increase the solubility of poorly water-soluble compounds [34].
`Hydrosols/nano-suspensions (note: from a physicochemi-
`cal classification point of view, crystalline and amorphous
`nano-suspensions belong to the group of hydrosols [8]) may
`also be considered for i.v. administration. However, as exem-
`plified in the Abraxane product, the preparation of amor-
`phous precipitates of palcitaxel and albumin as a matrix
`excipient will be very challenging at a small scale for NCEs
`at
`the preclinical research stage (see description of
`the
`Abraxane product below for more details).
`It is claimed that nano-suspensions with small particle size
`allowing i.v. administration are suitable to assess the absolute
`bioavailability [35]. This is an optimistic interpretation of the
`value of a nano-suspension injection because a prerequisite
`for determining absolute bioavailability is that the compound
`on i.v. injection is in the same molecular dispersed state in the
`blood as obtained after oral administration. It is, however, not
`clear to what extent the particles indeed do dissolve in the
`blood circulation, especially when it is claimed that the
`nanoparticles may have extended release properties on
`ingestion by macrophages [35]. The production of sterile,
`non-agglomerating nano-suspensions with well characterized
`crystal form and particle size distribution on a small scale
`may also be challenging. Oscillating beads milling equipment
`may be used for small scale milling [36]. The extrapolation of
`results obtained from oscillating bead milling equipment
`using only two exemplary compounds to poorly soluble
`compounds presented in this study may not be appropriate.
`
`1486
`
`Expert Opin. Drug Deliv. (2011) 8(11)
`
`

`

`A possible occurrence of amorphization or conversion to a
`metastable polymorph of the milled poorly water-soluble
`compounds has not been considered [36]. Also, the chemical
`stability of the compound during milling and sterility needs
`to be addressed. Finally, if there is a clear intention to use
`the identified preclinical research formulation for further
`development, it may be necessary to assess the IP freedom
`before embarking on preclinical testing with this technology.
`An example of a decision tree for selecting an i.v. formulation
`derived from [25] is provided in Figure 2.
`The formulation options are listed in decreasing degree of
`attractivity. The sequence of formulation options does not
`focus on the fact that every option has its pros and cons and
`that an adequate characterization of the drug substance before
`starting formulation screen should be a sound basis for the for-
`mulation screen. The characterization of the drug substance
`which, for example, encompasses pH-solubility and -stability
`profile and solubility profile in organic solvents would elimi-
`nate the first options of the formulation screen right away.
`For these reasons, Figure 3 summarizes a more objective
`approach, starting from a detailed drug substance characteriza-
`tion followed by in parallel assessment of the technologies,
`without taking preference for supposed simple approaches,
`but taking pros and cons of every option into consideration.
`Because the type of formulation may influence the i.v. PK
`(and in some cases even the body distribution) of the poorly
`water-soluble compound [37], it is important that toxicological
`assessments following the preclinical efficacy assessment will
`be performed with the same formulation (type) as intended
`for the clinic. In situations where there is a switch between
`the i.v. formulations of Figure 1, regulatory authorities may
`request, for example, bridging tox studies and repeating stabil-
`ity studies and QC development. For this reason, adequate
`attention should be paid during preclinical development to
`select the best i.v. formulation approach.
`
`3.2 Oral
`During early development stages, it is not possible to classify
`drugs intended for oral administration according to the bio-
`pharmaceutics classification system (BCS) [38] or to the
`recently proposed development classification system (DCS)
`[39] because initially the therapeutic dose is unknown. The
`DCS and BCS classify oral drug substances for solubility in
`water and permeation properties. For oral absorption of
`poorly water-soluble drugs according to the BCS, the mem-
`brane permeability kinetics could be either rapid or slow
`(BCS class II or IV, respectively). Although in vitro methods
`(Caco-2 and parallel artificial membrane permeation assay)
`can be used to assess these permeation kinetics, it is sometimes
`not possible due to lack of solubility in donor compartments
`for conclusive results.
`Li