OILY COMPOSITIONS

Abstract

The present invention provides emulsion compositions comprising an continuous oil phase, a discontinuous aqueous phase and a plurality of microparticles. The composition may comprise a pharmaceutical active ingredient located in the oil phase, the particulate phase or the aqueous phase. The emulsion compositions have improved stability and coherence and are useful in the treatment of tumours by embolotherapy.

Claims

1. An emulsion composition comprising a continuous phase, a discontinuous phase and a plurality of particles, the discontinuous phase being aqueous and the continuous phase comprising an oil; wherein the particles comprise a polymer to which iodine is covalently bound.

2. An emulsion composition comprising a continuous phase, a discontinuous phase and a plurality of particles, the discontinuous phase being aqueous and the continuous phase comprising an oil; wherein the particles are sufficiently hydrophobic such that an emulsion prepared according to example 2 herein, using a lipiodol:aqueous phase ratio of 2:1 and in which the aqueous phase contains no contrast agent, is stable for at least 10 minutes at between 18 and 22? C.

3. An emulsion composition comprising a continuous phase, a discontinuous phase and a plurality of particles, the discontinuous phase being aqueous and the continuous phase comprising an oil; wherein the particles, when measured according to Example 4a herein, have a cantilever deflection (measured in volts) of less than that of DC Bead.

4. A composition according claim 2 wherein the particles comprise a polymer to which iodine is covalently bound.

5. A composition according to claim 1 wherein the iodine is bound to an aromatic group, which aromatic group is covalently bound to the polymer.

6. A composition according to claim 1 wherein particles comprise a polymer to which iodine is covalently bound, and wherein the iodine is present in the particles at a level of at least 30 mg I/ml of packed volume, preferably 60 mg iodine per ml PV.

7. A composition according to claim 1 wherein the particles comprise a hydrogel polymer.

8. A composition according to claim 1 wherein the polymer carries an overall charge at a pH between 6 and 8.

9. A composition according to claim 8, wherein the polymer carries an anionic charge at a pH between 6 and 8.

10. A composition according to claim 1 wherein the volume of oil phase exceeds the volume of aqueous phase, preferably where the ratio of oil to aqueous phase is greater than 1.1:1 v/v.

11. A composition according to claim 1, which is stable for at least 10 minutes at between 18 and 22? C.

12. A composition according to claim 1 wherein the polymer comprises polyvinyl alcohol.

13. A composition according to claim 1 wherein the aqueous phase comprises a contrast agent.

14. A composition according to claim 1 wherein the particles comprise a pharmaceutical active ingredient.

15. A composition according to claim 1 where in the oil is a composition of iodised ethyl-esters of the fatty acids of poppy seed oil.

16. A process for preparing an emulsion according to claim 1 comprising: a. providing a continuous phase comprising an oil; b. providing an aqueous phase c. providing a plurality of particles which particles comprise iodine covalently bound to the particle; and combining them to provide an emulsion.

17. A process for preparing an emulsion according to claim 2 comprising: a. providing a continuous phase comprising an oil; b. providing an aqueous phase c. providing a plurality of particles; and combining them to provide an emulsion; wherein the particles provided are sufficiently hydrophobic such that: an emulsion prepared with those particles, according to example 2 herein, using a lipiodol:aqueous phase ratio of 2:1 and in which the aqueous phase contains no contrast agent, is stable for at least 10 minutes at between 18 and 22? C.

18. A process for preparing an emulsion according to claim 3 comprising: a. providing a continuous phase comprising an oil; b. providing an aqueous phase c. providing a plurality of particles; and combining them to provide an emulsion; wherein the particles provided, when measured according to Example 4a herein, have a cantilever deflection (measured in volts) of less than that of DC Bead.

19. An emulsion composition comprising a continuous phase, a discontinuous phase and a plurality of particles, the discontinuous phase being aqueous and the continuous phase comprising an oil; wherein either: a) the particle comprises a polymer to which iodine is covalently bound, b) the particles are sufficiently hydrophobic such that an emulsion prepared according to example 2 herein, using a lipiodol:aqueous phase ratio of 2:1 and in which the aqueous phase contains no contrast agent, is stable for at least 10, preferably at least 30, preferably at least 60, more preferably at least 80 and most preferably at least 90 minutes at between 18 and 22? C.; or c) the particles, when measured according to the protocol laid out in Example 4a, have a cantilever deflection (measured in volts) of less than that of DC Bead wherein the emulsion is preparable by a process according to claim 16.

20. A pharmaceutical active for use in the treatment of a tumour by embolotheraphy, wherein the pharmaceutical active is delivered, in an emulsion composition according to claim 1.

21. A method of embolotherapy in a patient having a tumour, comprising providing an embolic emulsion composition according to claim 1 and delivering the composition to blood vessels of the tumour.

22. A kit for preparing an emulsion comprising an oil and a plurality of particles, the particles comprising iodine which is covalently bound to the particle.

Description

FIGURES

[0098] FIG. 1 illustrates bead density and moisture content for beads of varying iodine content.

[0099] FIG. 2 illustrates emulsion stability and bead sedimentation for emulsions prepared with beads of varying iodine content using an emulsion of 2:1 oil to aqueous.

[0100] FIG. 3 compares photomicrographs of emulsions prepared using non iodinated beads with those prepared using beads having an iodine content of 155 mg I/ml. The figure is also representative particles at 33 mg/ml iodine.

[0101] FIG. 4 illustrates results of a saline drop test of representative emulsion preparations, low iodine particles had a 33 mg/ml iodine content and high iodine particles had a 155 mg/ml iodine content.

[0102] FIG. 5 is a diagrammatic representation of the vascular flow model used in examples.

[0103] FIG. 6 illustrates the scoring system for the behaviour of emulsions under flow conditions.

[0104] FIG. 7 shows illustrative high aqueous content and low aqueous content emulsions using high iodine (155 mg I/ml) and low iodine (33 mg I/ml) microspheres in the flow model.

[0105] FIG. 8 illustrates the effect of re emulsifying emulsions made with non iodinated and high iodine (147 mg I/ml) microspheres after partial separation of the emulsion.

[0106] FIG. 9 illustrates the elution profile of doxorubicin from emulsions of the invention in restricted flow conditions, modelling release of drug from embolic compositions in vivo.

[0107] FIG. 10 illustrates the results of atomic force microscopy studies on the microspheres. FIG. 9a illustrates the principle of calculating the pull-off (adhesion) force between the silicon tip and the microsphere surface and shows a force-distance curve with cantilever deflection. FIG. 9b illustrates the results obtained with an unmodified microsphere. FIG. 9c illustrates the results obtained using microspheres with 33 mg/ml iodine. FIG. 9d illustrates the results obtained using microspheres with 155 mg/ml iodine.

EXAMPLES

Example 1 Preparation of Iodinated Hydrogel Microspheres

[0108] A series of iodinated microspheres were prepared by coupling 2,3,5-triiodobenzaldehyde to preformed PVA AMPS hydrogel microspheres (DC Bead? Biocompatibles UK Ltd. Farnham: UK). The microspheres were prepared according to Example 1 of WO04071495, high AMPS version up to and including the step of vacuum drying to remove residual solvents. Coupling was then carried out according to Examples 5 and 6 of WO2015/033092, to provide microspheres that comprised triiodobenzyl groups linked via a cyclic acetal to the PVA backbone. Samples were prepared having an iodine content of between 33 and 155 mg/ml packed bead volume and a radiopacity of between 1020 and 6769 HU (measured according to Example 12 of WO2015/033092). Control, non iodinated beads were prepared according to Example 1 of WO04071495, high AMPS version and the process was continued to the end of the dying step. Microspheres were sieved to provide a size range of 70-150 um.

[0109] This iodine conjugation not only has the effect of increasing the radiopacity of the beads due to the presence of radio-dense iodine atoms, but also introduces hydrophobic moieties into the structure that reduce the water content and alter its surface properties.

[0110] Table 1 illustrates some physicochemical properties of this series of iodine-modified beads. FIG. 1 illustrates bead density and moisture content for beads of varying iodine content.

TABLE-US-00001 TABLE 1 Physicochemical properties of iodinated embolisation microspheres Iodine concentration (mg iodine per mL beads)* Characterisation 33 62 108 137 155 Solid Mass 99 190 288 363 384 (mg per mL beads) % Water Content 82 76 64 55 54 Radiopacity (HU) 1020 2120 3902 5395 6769 *Iodine content is determined per ml, packed volume, of fully hydrated microspheres in normal saline. Once lyophilised, the iodinated microspheres do not return to exactly the same volume on full rehydration, so all measurements of microsphere volume referred to herein refer to the iodine content per ml of fully hydrated microspheres before they have undergone lyophilisation.

Example 2: Preparation of Particle-Oil Emulsions

[0111] A series of emulsions preparations were made using either non-iodinated or iodinated microspheres (155 mg/ml iodine). Individual vials containing 2 ml of fully hydrated microspheres (packed volumemeasured by measuring cylinder) prepared according to Example 1, were lyophilised and stored dried under vacuum until used. Lipiodol? Ultrafluide (10 mL) was added to a vial of dry microspheres through the vial seal without breaking vacuum, via syringe needle, and mixed well for 5 minutes to ensure uptake of the oil into the microspheres. The microspheres were then transferred into a 20 mL polypropylene syringe (Becton Dickson). Two milliliters of an aqueous doxorubicin solution (25 mg/mL) was aspirated into a 10 mL polypropylene syringe (Becton Dickson). Additional water for injection or Omnipaque? 350 contrast agent (GE Healthcare) was added to the doxorubicin solution as required, to obtain the required ratio of oil and aqueous phase when mixed together with the microspheres.

[0112] The syringe containing the doxorubicin mixture was attached to the 20 mL syringe containing the microspheres and Lipiodol, using a polyamide 3 way stop-cock (Discofix? B.Braun). Initial mixing of the 2 syringe contents was performed by slowly adding the aqueous drug solution to the Lipiodol and microspheres and mixing in order to generate droplets of aqueous drug in the oil phase and to form a homogenous mixture.

[0113] The contents were then mixed rapidly between syringes (20 times). This was repeated every 5 minutes for 30 minutes to ensure partition of the drug into and suspension of the microspheres in the emulsion.

[0114] The emulsion formulations used are given in Table 3.

Example 3: Evaluation of Particle-Oil Emulsions

[0115] 3a. Stability:

[0116] The prepared particle-oil emulsions were transferred into a 10 mL polypropylene syringe (Becton Dickson) and placed in an upright orientation i.e. the length of the syringe in a vertical position. The separation of the oil and the aqueous phase, and the sedimentation of the microspheres in the emulsion were visually monitored. The emulsion stability time was determined as the time taken for separation of the oil and aqueous phases reaching 10% of the initial emulsion volume. In formulations showing settling of microspheres, the time taken for the sedimentation of microspheres to reach approximately 10% of the initial emulsion volume was also noted.

[0117] Emulsions with various ratios of oil and water were evaluated for their stability using the most hydrophobic microsphere (155 mg/mL iodine) and the hydrophilic microsphere (no iodine). Table 3 reports the comparative stability of emulsions assessed according to the criteria laid out in Table 2.

TABLE-US-00002 TABLE 2 Assessment Rating Criteria for Emulsion stability Emulsion Stability Microsphere Rating (minutes) Sedimentation (minutes) + <1 <1 + + 2-3 2-3 + + + 3-5 3-5 + + + + 5-10 5-10 + + + + + >10 >10

[0118] The comparative stability of emulsions containing the hydrophobic microspheres in oil-water ratios ranging from 1:1 to 5:1 was much better than that observed in emulsions with the hydrophilic microspheres (ie no iodine), which were typically only stable for between 2-3 minutes.

[0119] Emulsion stability and bead sedimentation in relation to bead iodine content for a representative emulsion with an oil to aqueous ratio of 2:1 (66.7% Lipiodol, 33.3% aq of which 20% was contrast) are illustrated in FIG. 2.

TABLE-US-00003 TABLE 3 Emulsion Stability Assessment Emulsion Component Stability Rating Composition (%) with Microspheres: Approximate Aqueous Hydrophobic Oil:Water Contrast Hydro- (155 mg/mL ratio Lipiodol Total Agent philic iodinated) 1 1:1 52.6 47.4 15.8 Not tested +++++ 2 1.5:1 58.8 41.2 17.7 Not tested +++++ 3 1.5:1 58.8 41.2 29.4 +++ +++++ 4 1.5:1 58.8 41.2 38.2 Not tested +++++ 5 1.5:1 62.5 37.5 0.0 Not tested +++++ 6 2:1 66.7 33.3 0.0 Not tested +++++ 7 2:1 66.7 33.3 20.0 ++ +++++ 8 2.5:1 71.4 28.6 3.6 ++ +++++ 9 5:1 83.3 16.7 12.5 Not tested +++++ 10 5:1 83.3 16.7 0.0 Not tested +++++ 11 2:1 66.7 33.3 25.0 ++ Not tested 12 2.2:1 68.9 31.1 17.2 ++ Not tested 13 2.5:1 71.4 28.6 14.3 + Not tested 14 2.5:1 71.4 28.6 21.4 + Not tested 15 3:1 74.1 25.9 11.1 + Not tested

[0120] 3b. Appearance:

[0121] A drop of emulsion containing either non iodinated or iodinated microspheres (155 or 33 mg/ml iodine) was placed onto a petri dish and immediately covered with a glass cover slip. Optical micrographs (?4 or ?10 magnification) were obtained using a BX50 microscope, Colorview III Camera and Stream Essential imaging software (Olympus). In order to differentiate between the oil and aqueous phase, an aqueous solution of Reactive Blue 4 dye in water (50 mg/mL) was added dropwise to the emulsion on the petri dish. The aqueous phase of the emulsion could be identified by the migration of the blue dye solution towards it under the microscope. The type of emulsion (oil-in-water or a water-in-oil) and the location of the microspheres, in relation to the oil or aqueous phases, were noted.

[0122] FIG. 3 illustrates particle-oil emulsions. The iodinated microspheres have a preference to reside in the oil phase despite their relatively high water content, and accumulate at the liquid droplet interface, hence stabilising the droplets from coalescence. Even at the 1:1 ratio of oil to aqueous, where the continuous phase of the emulsion appears to be the aqueous, the iodinated microspheres are still predominantly populated around the oil droplets and reside at the interface between the water and oil. The non-iodinated microspheres are more hydrophilic in comparison and are observed to reside in the aqueous phase, even in the emulsions with a high oil composition (i.e. 2:1 and 3:1 oil-water ratio).

[0123] 3c. Emulsion Stability (Cohesiveness) in Static Flow Conditions (Saline Droplet Test)

[0124] A sample of emulsion was delivered below the surface of a 0.9% saline solution through an 18 gauge blunt fill needle (Becton Dickson) at room temperature. The appearance and behaviour of the emulsion, was observed and rated according to the characteristics defined in Table 4. The emulsions appearing as spherical droplets containing the microspheres without fragmenting or disintegrating were considered to have the better flow behaviour and emulsion stability.

TABLE-US-00004 TABLE 4 Assessment Rating Criteria for Flow Characteristics in Static Conditions Rating Emulsion Characteristic in static flow + Loose beads separated from oil and residing in the aqueous phase i.e. no association of beads with oil. + + Stream of oil with some beads associated with oil. Beads mainly residing in the aqueous phase (saline). + + + Stream of oil where beads are associated with the oil. Forms droplets with beads significantly at the oil/aqueous interface. Beads may aggregate towards the aqueous. + + + + Stable oil droplets containing beads. Some streams of oil and beads combined - beads within the stream are visually bound together by the oil. + + + + + Stable oil droplets in which the beads reside are formed consistently. Oil droplets hold together.

[0125] FIG. 4 illustrates the flow characteristics observed in representative particle-oil emulsions having oil to water ratios ranging from approximately 1.5:1 to 5:1. The more hydrophobic microspheres containing iodine resulted in droplets that were more spherical and had better flow characteristics than the comparatively hydrophilic microspheres containing no iodine. Increasing aqueous content in the particle-oil emulsions resulted in droplets that were less spherical. This was more apparent in microspheres containing no iodine. Table 5 illustrates the results

TABLE-US-00005 TABLE 5 Behaviour of emulsions containing hydrophobic microspheres (33 and 155 mg/mL) and hydrophilic microspheres under static flow Component Ap- Composition (%) Emulsion Flow Rating with proximate Aqueous Microspheres: Oil:Water Contrast Hydro- Hydrophobic* ratio Lipiodol Total Agent philic (iodinated) 1 1:1 52.6 47.4 15.8 Not tested +++ 2 1.5:1 58.8 41.2 17.7 Not tested +++ 3 1.5:1 58.8 41.2 29.4 + +++? 4 1.5:1 58.8 41.2 38.2 Not tested ++++ 5 1.5:1 62.5 37.5 0.0 Not tested ++++ 6 2:1 66.7 33.3 0.0 Not tested +++++ 7 2:1 66.7 33.3 20.0 ++ +++++ 8 2.5:1 71.4 28.6 3.6 ++ +++++ 9 5:1 83.3 16.7 12.5 Not tested +++++ 10 5:1 83.3 16.7 0.0 Not tested +++++ 11 2:1 66.7 33.3 25.0 ++ Not tested 12 2.2:1 68.9 31.1 17.2 +++ Not tested 13 2.5:1 71.4 28.6 14.3 +++ Not tested 14 2.5:1 71.4 28.6 21.4 +++ Not tested 15 3:1 74.1 25.9 11.1 +++ Not tested *Microspheres having 33 mg I/ml and 155 mg I/ml behave the same.

[0126] Emulsions containing iodinated microspheres at both 33 mg/mL and 155 mg/mL levels, formed stable oil droplets containing the microspheres, in compositions where the oil:water ratio was at least 2:1. At lower oil-water ratios. the emulsion was still able to form stable oil droplets containing the particles.

[0127] In comparison, the microspheres containing no iodine showed flow behaviour that was poor in comparison to the iodinated samples. Additional formulations with these microspheres were tested, with oil-water ratios ranging from 2:1 to 3:1 and varying contrast agent compositions. These did not improve the emulsion flow behaviour such that it was comparable to that observed with hydrophobically modified microspheres.

[0128] 3d. Behaviour Under Continual Flow

[0129] An in vitro vascular flow simulator (FIG. 5) was used to profile the physical stability of emulsions under continual flow conditions. The particle-oil emulsions tested under static flow conditions (in c. above) were tested under non-restricted continual flow conditions.

[0130] The emulsions were transferred into a 3 mL polypropylene syringe (Becton Dickson) attached to microcatheter (Progreat 2.4Fr, Terumo) for administration into a silicone vascular flow simulator model (Elastrat Sarl, Switzerland) set-up as an open loop system.

[0131] The distal end of the microcatheter was located proximally in the vascular channels and a 0.9% saline solution at ambient temperature was pumped through the flow model at a pulsatile rate of 60 mL/minute, by a peristaltic pump. The particle-oil emulsion was administered manually through the microcatheter at an injection rate of approximately 0.5 mL/minute. The appearance of the emulsion delivered from the microcatheter and into the vascular flow model and delivery channels was observed for its flow characteristics (Table 5) and was graded for the ability to form discreet oil droplets and maintain the droplet integrity during its flow through the in vitro vascular network. The emulsions appearing as discreet spherical oil droplets containing the microspheres were considered to have the best flow behaviour and stability i.e. high scoring, whereas streaming beads (not within the oil droplet) were defined as being less stable and low scoring.

[0132] The minimum requirement for acceptable stability being that the emulsion had a rating of (+++) i.e. appeared as a stream of oil and beads together or as stable droplets or a stream of oil containing beads which are significantly at the oil/aqueous interface.

TABLE-US-00006 TABLE 6 Assessment Rating Criteria for Flow Characteristics in Continuous flow Score Emulsion Characteristic in continual flow (open loop) + Stream of loose beads. Oil and beads completely separate. Immediate drop out of beads from oil on delivery. + + Some stream of oil and beads together. A few beads at the oil/aqueous interface but beads drop out under flow. + + + Stream of oil and beads together. Stable droplets or stream of oil containing beads which are significantly at the oil/aqueous interface. Beads may drop out of the interface during passage through bifurcation + + + + Mainly stable droplets formed or solid stream of oil with beads residing in the oil phase. Good retention of beads with no drop out during passage through bifurcation. + + + + + Stable droplets formed consistently. Beads residing in the oil phase. Good retention of beads with no drop out during passage through bifurcation.

[0133] FIG. 7 illustrates example emulsions in flow conditions and shows that under continual flow, the emulsions prepared with the microspheres containing iodine consistently produce stable droplets in which the microspheres reside, with complete retention as the droplets pass through the bifurcations of the vascular model.

[0134] Increasing aqueous content in the particle-oil emulsions has a direct impact on their flow characteristics with the non iodine microspheres not able to form stable droplets of oil, containing the microspheres, even when comprising a high oil content. In comparison, the hydrophilic microspheres did not form robust droplets containing beads. In emulsion compositions with high oil content, oil droplets did form but the microspheres tended to reside at the interface of the oil and had a tendency to be displaced and fall out under continual flow.

[0135] Table 7 reports the assessment of flow characteristics according to Table 5.

TABLE-US-00007 TABLE 7 Behaviour of emulsions containing hydrophobic microspheres (33 and 155 mg/mL) under continual flow Component Emulsion Composition (%) Flow Rating with Approximate Aqueous Microspheres: Oil:Water Contrast Hydro- Hydrophobic ratio Lipiodol Total Agent philic (iodinated) 1 1:1 52.6 47.4 15.8 Not tested +++ 2 1.5:1 58.8 41.2 17.7 Not tested ++++ 3 1.5:1 58.8 41.2 29.4 + +++++ 4 1.5:1 58.8 41.2 38.2 Not tested +++++ 5 1.5:1 62.5 37.5 0.0 Not tested +++++ 6 2:1 66.7 33.3 0.0 Not tested +++++ 7 2:1 66.7 33.3 20.0 ++ +++++ 8 2.5:1 71.4 28.6 3.6 +++ +++++ 9 5:1 83.3 16.7 12.5 Not tested +++++ 10 5:1 83.3 16.7 0.0 Not tested +++++ 11 2:1 66.7 33.3 25.0 ++ Not tested 12 2.2:1 68.9 31.1 17.2 +++ Not tested 13 2.5:1 71.4 28.6 14.3 +++ Not tested 14 2.5:1 71.4 28.6 21.4 +++ Not tested 15 3:1 74.1 25.9 11.1 +++ Not tested

[0136] These findings are supported in vivo following fluoroscopic guided delivery in healthy swine and evaluation of flow properties. Iodinated microspheres (XXmg I/ml ?70-150 um) flowed as discreet packets of emulsion as per the in vitro conclusions, whilst cTACE Lipiodol emulsions and emulsions prepared with unmodified microspheres presented loose whispy flow properties indicative of poor flow stability.

[0137] 3e. Effect of Aqueous Phase Density on Particle-Oil Emulsions

[0138] The behaviour of the particle-oil emulsion when the aqueous phase is water (with a much lower specific gravity than that of the contrast agent) is comparable to emulsions in which the aqueous phase comprises contrast agent. FIG. 8 shows representative images of the flow characteristics observed in particle-oil emulsions which were prepared using either water or contrast agent, in an oil-water ratio of 2:1 (Table 8).

TABLE-US-00008 TABLE 8 Composition of Emulsion % Contrast Agent component % Lipiodol % Aqueous (in Aqueous) 66.7 33.3 20.0 66.7 33.3 0

[0139] The homogeneity of the emulsion droplet in the static flow test (saline drop) was slightly less in the emulsion prepared with water only, however, under continual flow it was comparable to the emulsion prepared with contrast agent.

[0140] 3f. Emulsion Characteristics on Re-Mixing

[0141] The particle-oil emulsions were evaluated for their flow characteristics and their appearance after initial preparation and following phase separation and subsequent re-mixing. Particle-oil emulsions containing the more hydrophilic microspheres, having no bound iodine, do not form stable and robust emulsion under flow or in the syringe. Once phase separation has occurred (3 minutes), re-mixing of the emulsion exacerbated the separation and overall syringe stability was reduced further to 1 minute.

[0142] In contrast, the iodine containing microspheres formed an emulsion which was stable for over 60 minutes and once re-mixed was still stable in the syringe and able to form stable droplets under continual flow. On phase separation and coalescence of the oil and aqueous droplets, the hydrophobic microspheres are capable of stabilising the emulsion after re-mixing, and demonstrating their original characteristics (see FIG. 8).

[0143] 3g. In Vitro Drug Elution Behaviour of Emulsions in a Vascular Flow Simulator with Restricted Flow

[0144] The drug release front the particle oil emulsions was profiled under continual flow conditions with a partially restricted flow rate on sample delivery. The particulate-oil emulsions and in vitro system were prepared as in Section 3d but with the in vitro system set-up as a semi closed-loop system as follows.

[0145] A single sample delivery channel was restricted using a 27 ?m microporous mesh (woven polyamide) on its outlet port to simulate the effect of flow confinement in micro-channels once the emulsion was administered. The other outlets on the silicone vascular phantom, which were not used for sample delivery, were left unrestricted, with their outlets recirculating the elution media (0.9%, saline solution) back into the saline reservoir (heated to 37? C.). The microcatheter was placed distally into the single sample delivery channel and the saline solution was pumped through the restricted sample delivery channel at a flow rate of approximately 80 mL/minute.

[0146] The particle-oil emulsion prepared as above were administered through the microcatheter using a syringe auto-injector (PHD Ultra; Harvard Apparatus) at an injection rate of 0.5 mL/minute. The emulsions were delivered until the flow rate at the exit port of the delivery channel was observed to have significantly reduced. During the delivery of the emulsion, the saline from the sample delivery channel was captured and analysed by UV/Vis spectrophotometry (Cary 50 Bio, Varian) at a wavelength of 483 nm to determine the amount of doxorubicin eluted. The cumulative drug eluted over a 20 min time period, was normalised by determining it as a percentage of the theoretical dose administered. Table 9 give the details of the emulsions for which data is shown in FIG. 9.

TABLE-US-00009 TABLE 9 Composition of emulsions tested in Example 3g of which Total Contrast Iodine O:W Ratio Lipiodol % Aqueous % Agent is details Article 1 155 5:1 83.3 16.7 0.0 10 mL oil + 2 mL Low dox in water Aqueous Article 2 0 5:1 83.3 16.7 0.0 10 mL oil + 2 mL Low dox in water Aqueous cTACE 5:1 83.3 16.7 0.0 10 mL oil + 2 mL dox in water

[0147] 4a. Measurement of Microsphere Surface Properties

[0148] Relative hydrophilicity/hydrophobicity of the microsphere surfaces was measured by use of Contact Atomic Force Spectroscopy using force-distance-amplitude protocols.

[0149] Initial qualitative assessments were performed on a Dimension 3000 atomic force microscope (AFM) (Digital Instruments) and data collected using the NanoScope IIIA software (Digital Instruments). The force distance curve was obtained by measuring the deflection signal (voltage) from the cantilever as the probe approached the surface from approximately 750-1800 nm above the surface at a constant scan rate of 1 Hz. The cantilever deflection was observed as the difference in signal voltage at the points when the tip is retracting from the microsphere surface and when the tip is free from the surface (FIG. 10). Using the silicon AFM probe (PointProbe NCH-W; Nanosensors), it was expected that the adhesion and cantilever deflection would be highest in microspheres having a more hydrophilic surface property.

[0150] The microsphere containing no iodine has the highest cantilever deflection signal of 0.8V and the lowest signal of 0.1V was observed in the microsphere with the highest amount of iodine (Table 10), thus indicating comparative hydrophobicity increases with increasing iodine in the microspheres.

TABLE-US-00010 TABLE 10 AFM Cantilever Deflection Signals in Microspheres of Differing Iodine Content Iodine content (mg/mL) Cantilever Deflection (V) 0 0.8 33 0.2 155 0.1

[0151] 4b Alternate AFM Measurement

[0152] Further verification of the adhesion force was performed using an Asylum Research Cypher AFM instrument (Oxford Instruments) and a CONTSCR-10 cantilever (silicon probe) (Nanoworld). A force map was generated for each microsphere within an overall scan size area of 4 ?m by 4 ?m. The force map comprised a total of 256 individual force distance curves (within the scan area) which were taken 250 ?m apart from each other (i.e. in a regular array of 16?16 measurements).

[0153] The force-distance curves were obtained by measuring the deflection of the AFM cantilever (holding an uncoated silicon probe) as it approached and retracted from the microsphere surface. Each curve was taken at a constant scan rate of 1 Hz. and comprised of 2000 points during the approach and retract cycle. The tip of the probe started 300 nm above the microsphere and was pressed 20 nm into the microsphere surface before being retracted. Depending on the curvature of the microsphere surface, the z height (distance of the AFM probe tip above the surface of the microsphere) over the area scanned was approximately 300-800 nm. The cantilever deflection was measured as a signal voltage and was calculated to give an adhesion force using the spring constant of the cantilever.

[0154] These assessments to determine the adhesion force on the hydrophobic microspheres (Table 11) show that the hydrophobic microsphere with the lower iodine concentration has the largest pull-off force and therefore the greater adhesion with the silicon AFM probe. Since the probe is considered to be hydrophilic, the lower iodine sample is determined to be less hydrophobic than the higher iodine sample.

TABLE-US-00011 TABLE 11 AFM Adhesion Forces in Microspheres Containing Iodine Average pull-off force (nN) Sample Replicate 1 Replicate 2 Low Iodine Microsphere 11.6 ? 8.8 8.6 ? 4.7 (33 mg/mL) High Iodine Microsphere 4.1 ? 1.4 4.4 ? 1.5 (155 mg/mL)