Process for preparation of beads for imaging

Abstract

A process for the preparation of beads including a biocompatible hydrophobic polymer, a perfluorocarbon, polyvinylalcohol and optionally a metal compound, including the steps of: adding the perfluorocarbon and optionally the metal compound to a solution of the biocompatible hydrophobic polymer in a polar solvent to provide a first liquid mixture, adding the first liquid mixture to an aqueous solution of a biocompatible surfactant including polyvinylalcohol under sonication to obtain a second liquid mixture, a) maintaining the sonication of the second liquid mixture while cooling, b) evaporating the polar solvent from the second liquid mixture to obtain a suspension of beads including the biocompatible hydrophobic polymer, the perfluorocarbon and optionally the metal compound, c) separating the beads from the suspension and preparing a water suspension of the beads and d) freeze-drying the water suspension to obtain the beads, wherein the addition of the first liquid mixture to the biocompatible surfactant in step b) is performed within a period of at most 10 seconds, wherein the sonication in step b) and the sonication in step c) are performed directly into the liquid mixtures by for example a probe or flow sonicator at an amplitude of at least 120 μm for 0.01-10 minutes and wherein the weight ratio of the biocompatible surfactant to the biocompatible hydrophobic polymer is at least 3:1. Beads having close F—H2O interactions, which are suitable for imaging purposes.

Claims

1. A process for the preparation of beads having a multi-domain structure comprising a biocompatible hydrophobic polymer, a perfluorocarbon, polyvinylalcohol (PVA) and optionally a metal compound, comprising the steps of: a) adding the perfluorocarbon and optionally the metal compound to a solution of the biocompatible hydrophobic polymer in a polar solvent to provide a first liquid mixture; b) adding the first liquid mixture to an aqueous solution including the PVA as a first biocompatible surfactant and optionally a second biocompatible surfactant under sonication to obtain a second liquid mixture; c) maintaining the sonication of the second liquid mixture while cooling; d) evaporating the polar solvent from the second liquid mixture to obtain a suspension of beads comprising the biocompatible hydrophobic polymer, the perfluorocarbon and optionally the metal compound; e) separating the beads from the suspension and preparing a water suspension of the beads; and f) freeze-drying the water suspension to obtain the beads, wherein the beads have multiple perfluorocarbon domains distributed in a matrix comprising the biocompatible hydrophobic polymer and the PVA; wherein the addition of the first liquid mixture to the aqueous solution in step b) is performed within a period of at most 10 seconds, wherein the sonication in step b) and the sonication in step c) are performed directly to the liquid mixtures at an amplitude of at least 120 μm for 0.01-10 minutes and wherein the weight ratio of the PVA and optionally the second biocompatible surfactant to the biocompatible hydrophobic polymer is at least 3:1.

2. The process according to claim 1, wherein step a) involves adding the perfluorocarbon and the metal compound to the solution of the biocompatible hydrophobic polymer, wherein the metal compound is added without prior dilution or as a solution comprising at least 100 mg of the metal compound per mL of the solution and wherein the first liquid mixture obtained is an emulsion.

3. The process according to claim 1, wherein step a) is performed under direct sonication.

4. The process according to claim 1, wherein the biocompatible hydrophobic polymer comprises a polymer selected from the group consisting of poly(lactic-co-glycolic) acid, polylactic acid), poly(caprolactone), polydimethylsiloxane and combinations thereof.

5. The process according to claim 1, wherein the polar solvent is selected from the group consisting of dichloromethane, chloroform, ethyl acetate and combinations thereof.

6. The process according to claim 1, wherein the perfluorocarbon is selected from the group consisting of perfluoro crown ether, perfluoro octyl bromide, perfluorooctane, perfluoro poly ethers and combinations and modifications thereof.

7. The process according to claim 1, wherein the metal compound comprises gadolinium.

8. The process according to claim 1, wherein the second biocompatible surfactant is present and is selected from the group consisting of a polysorbate and polyvinylpyrrolidone.

9. The process according to claim 1, wherein the weight ratio of the perfluorocarbon in the first liquid mixture with respect to the biocompatible hydrophobic polymer in the first liquid mixture is between 10:1 to 25:1, or between 12:1 to 20:1, or between 14:1 to 18:1.

10. The process according to claim 1, wherein the weight ratio of the metal compound in the first liquid mixture with respect to the biocompatible hydrophobic polymer in the first liquid mixture is between 1:1000 to 1:100, or between 1:500 to 1:150, or between 1:300 to 1:180.

11. The process according to claim 1, wherein the weight ratio of the biocompatible surfactant with respect to the biocompatible hydrophobic polymer is between 4:1 to 10:1, or between 4.5:1 to 8:1 or between 4.8:1 to 6:1.

12. The process according to claim 1, wherein step a) is performed under direct sonication, using a probe.

13. The process according to claim 1, wherein step a) is performed under direct sonication, using a flow sonicator.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows 1H NMR spectra of particles produced by the process of WO2014/041150 (left) and new (right) process. The spectra were measured with resorcinol (4.1 mg) as an internal reference; 400 MHz, D.sub.2O, relaxation decay=20 s. Inset in the right spectrum: chemical structure of PVA, non-hydrolyzed OH groups not shown for simplification.

(2) FIG. 2 shows the DSC curves measured on beads comprising PLGA, PFCE and gadoteridol according to the invention. No melting of PFCE at −17° C. and no crystallization at −19° C. can be detected indicating that PFCE phase does not have a distinct melting point. Heating rate 2 K/min, sample initial weight 10.3 mg, sample final weight 5.6 mg.

(3) FIG. 3 shows Hetero-nuclear Overhauser Enhancement Spectroscopy (HOESY) NMR measured on beads comprising PLGA and PFCE without gadoteridol according to the invention. HOESY between water-protons and fluorine can be detected. This indicates a homogeneous structure with small sized perfluorocarbon domains distributed in the polymeric matrix. Nanoparticles in D.sub.2O, 500 MHz (.sup.1H), 471 MHz (.sup.19F).

(4) FIG. 4 shows solid state NMR measured on beads comprising PLGA, PFCE and gadoteridol according to the invention, which are in a dry state and a water-swollen state. FIG. 4 shows spin-lattice relaxation time T1 measured by solid state NMR (10 kHz MAS). There are differences in the spin-lattice relaxation time between the dry nanoparticles indicated by circles and the nanoparticles swollen with water indicated by squares. This shows that water can freely enter the entire structure, again indicating a homogeneous structure that contains small sized perfluorocarbon domains, which are distributed between polymeric chains.

(5) FIG. 5 shows solid state NMR measured on beads comprising PLGA, PFCE and gadoteridol according to the invention, which are in a dry state and a water-swollen state. FIG. 5 shows spin-spin relaxation time T2 measured at solid state NMR (10 kHz MAS). There are differences in the spin-lattice relaxation time between the dry nanoparticles indicated by green circles and the nanoparticles swollen with water indicated by blue squares.

(6) FIG. 6 shows small angle X-ray scattering results of the beads comprising PLGA, PFCE and gadoteridol (line with higher intensity) and the beads comprising PLGA without PFCE and gadoteridol (line with lower intensity). The results are similar, which may, suggest that the particles are homogeneous full sphere or particle with polydisperse PFCE-filled pores.

(7) FIG. 7 shows X-ray diffraction results of the beads comprising PLGA, PFCE and gadoteridol (line with lower intensity) and the beads comprising PLGA without PFCE and gadoteridol (line with higher intensity). The results are similar.

(8) FIG. 8 shows results of thermogravimetric (TGA) analysis of the beads comprising PLGA, PFCE and gadoteridol according to the invention, PLGA and PGA. The shoulder at 400° C. in the result of the beads according to the invention indicates the presence of PVA in the beads.

(9) FIG. 9 shows HOESY measurement of experiment 15 (comparative). HOESY NMR of PFCE-loaded core-shell capsules in D2O. The cross-peak between water and PFCE (approx. −92 ppm y-axis 4.8 ppm x-axis) is almost not present, showing that fluorine does not interact with water. The residual signal is coming from the region where the PFCE core is in contact with the shell. 400 MHz (.sup.1H), 377 MHz (.sup.19F).

(10) FIG. 10 shows the relation between amplitude and power on sonicators having different microtips. A microtip of 3.2 mm gives the highest amplitude at a certain amplitude setting, followed by a microtip of 4.8 and 6.5 mm respectively.

DETAILED DESCRIPTION OF THE INVENTION

(11) The invention is now elucidated by way of the following examples, without however being limited thereto.

EXAMPLES

Experiments 1-6: Preparation of Beads Comprising a Metal Compound

(12) PLGA (100 mg, resomer 502H) was dissolved in 3 mL dichloromethane. Perfluoro-15-crown-5 ether (900 μL, 1600 mg) and Prohance (1.78 mL, 497 mg gadoteridol) were added to the solution of PLGA and a first emulsion was formed by sonication using a microtip having a tip diameter of 3 mm at an amplitude of 40% for 15 seconds (Digital Sonifier s250 from Branson). This first emulsion was rapidly (within 10 seconds) added to a solution of poly(vinyl alcohol) (9000-10000 Da Mw and 80% hydrolysis) (25 g of water and 100-500 mg of PVA) in a round bottom flask while sonication of PVA-containing flask was started. The entire mixture was sonicated in ice-water bath using a microtip having a tip diameter of 3 mm at an amplitude of 20% or 40% to obtain a second emulsion. A sonication setting of 20% refers to an amplitude of about 160 μm, while a sonication setting of 40% relates to an amplitude of ca 300 μm.

(13) The duration of the period from the addition of the first emulsion to the end of the sonication was 3 minutes (Digital Sonifier s250 from Branson).

(14) After sonication dichloromethane was evaporated at 4° C. or room temperature overnight under stirring to achieve solidification of the beads. The beads were isolated by centrifugation at 27200 g for 20 min in 50 mL centrifugation tubes and resuspended in 25 g of water. The washing step was repeated two more times with resuspention by sonication after second washing (sonication bath, Diagenode Bioruptor). After washing, beads were resuspended in 4 mL of water, frozen with liquid N.sub.2 and freeze-dried. The resulting product was a white powder.

(15) The amounts of the components and the sonication amplitude which were varied are shown in Table 1, together with the properties and the yield of the beads.

(16) TABLE-US-00001 TABLE 1 Radius PFCE- Exp. PVA/ Sonication (DLS; content/ yield/ No mg Amplitude intensity)/nm PDI wt.-% mg 1 100 20% 357 0.49 11 77 2 500 20% 121 0.1 5.3 55 3 100 40% 314 0.39 28 137 4 200 40% 174 0.2 34 189 5 350 40% 146 0.15 39 184 6 500 40% 121 0.123 45 204

(17) Small beads with narrow particle size distribution were obtained by the process according to the invention (Ex 4, 5 and 6). It can be observed that a high amplitude (40%) and a large amount of PVA (200 mg (8.3 wt. %), 350 mg (13.7 wt. %) or 500 mg (18.5 wt. %)) resulted in a desirable combination of a small radius, low PDI, a high PFCE content and a high yield.

Experiments 7-9: Preparation of Beads Without a Metal Compound

(18) PLGA (100 mg, resomer 502 H) was dissolved in 3 mL dichloromethane (DCM) followed by addition of perfluoro-15-crown-5 ether (900 μL). The resulting double phase liquid was rapidly added with a glass pipette to a solution of poly(vinyl alcohol) (25 g of water and 100-500 mg of PVA) in a round bottom flask while sonication was started. Care was taken so that the phase of PLGA/DCM and the phase of PFCE were added simultaneously at a constant ratio. The entire mixture was sonicated in ice-water bath using a microtip having a tip diameter of 3 mm at an amplitude of 20% or 40% to obtain an emulsion. The duration of the period from the addition of the double phase liquid to the end of the sonication was 3 minutes (Digital Sonifier s250 from Branson).

(19) After sonication dichloromethane was evaporated at 4° C. or room temperature overnight under stirring to achieve solidification of the beads. The beads were isolated by centrifugation at 27200 g for 35 min in 50 mL centrifugation tubes and resuspended in 25 g of water. The washing step was repeated two more times with resuspention by sonication after second washing (sonication bath, Diagenode Bioruptor). After washing, beads were resuspended in 4 mL of water, frozen with liquid N.sub.2 and freeze-dried. The resulting product was a white powder with a yield of at least 100 mg.

(20) TABLE-US-00002 TABLE 2 Radius PFCE- Exp. PVA/ Sonication (DLS; content/ yield/ No mg Amplitude intensity/nm PDI wt.-% mg 7 100 20% 354 0.25 15 86 8 100 40% 339 0.24 25 116 9 500 40% 100 0.04 48 154

(21) Small beads with narrow particle size distribution and a high PFCE content were obtained with a high yield according to the process of the invention (Ex 9).

Experiments 10-12 (Comparative)

(22) Experiment 6 was repeated except that the PLGA was dissolved in a solvent indicated in Table 3.

(23) TABLE-US-00003 TABLE 3 Sonication Radius PFEC- Exp. PVA/ Ampli- (DLS; content/ yield/ No Solvent mg tude intensity)/nm PDI wt.-% mg 10 THF 500 40% 171 0.5 15 131 11 Acetone 500 40% 294 0.66 8 90 12 Aceto- 500 40% 223 0.22 5 95 nitrile

(24) Beads obtained are larger and have a broader size distribution than the experiments in which the solvent was dichloromethane.

(25) Diameter of beads prepared according to examples 1-12 was determined using dynamic light scattering (DLS) as described in Biomaterials. 2010 September; 31 (27):7070-7.

Experiment 13 (Comparative): Preparation of Beads Using Cup Horn

(26) PLGA (90 mg, resomer 502 H) was dissolved in 3 mL dichloromethane. Perfluoro-15-crown-5 ether (890 μL) was added to the solution of PLGA. 50 mL of an aqueous solution comprising of Prohance with concentration of 3 mg/mL was further added. This mixture was added dropwise to a solution of poly(vinyl alcohol) (20 g/L) in a glass tube while sonication of PVA-containing flask was started. The entire mixture was sonicated in a cup horn at an amplitude of 30% for 3 minutes, with 60 s on and 10 s off cycles (Digital Sonifier s250 from Branson) to obtain a second emulsion. During the sonication the temperature of the cooling water was maintained at 4° C. by a refrigerated circulator.

(27) After sonication dichloromethane was evaporated at 4° C. overnight under stirring to achieve solidification of the beads. The beads were isolated by centrifugation at 21000 g for 30 min in 2 mL centrifugation tubes and resuspended in 25 g of water. The pellet was washed with water twice and then resuspended in water, frozen at −80° C. and freeze-dried. The resulting product was a white powder with a yield of 50 mg.

(28) The examples according to the invention (Ex 5, 6, 9) resulted in a much higher yield compared to experiment 13.

Experiment 14 (Comparative); Use of Other Surfactants (No PVA)

(29) To study whether PVA is necessary for the formation of the particles of the invention, we tested the production with other surfactants that are commonly used for production of empty PLGA particles without perfluorocarbon (table 4). However, with all these surfactant we observed strong increase in size and polydispersity of nanoparticles. Especially, the PDI values, which are all higher than 0.5, demonstrate the very broad size distribution of the samples. Moreover, the encapsulation of PFCE was significantly lower or even not measurable with NMR. In summary the synthesis of nanoparticles was not possible with other surfactants. Therefore, we conclude that PVA is essential for stabilization of nanoparticles and for encapsulation of PFCE.

(30) TABLE-US-00004 TABLE 4 Production of nanoparticles with different surfactants. Surfactant R.sub.h/nm PDI PFCE-content/wt.-% Tween 20 215 0.81 n/a Sodium cholate 390 0.85 6 poly (vinylpyrrolidone) 450 0.77 3 Pluronic F68 315 0.45 n/a

TEST METHODS

Quantification of PFCE

(31) PFCE content was measured with 19F NMR at Bruker Avance III 400 MHz spectrometer using D2O (sigma-aldrich) as solvent and trifluoroacetic acid as an internal reference. The relaxation delay D1 was set to 20 s. Data evaluation was done with MestreNova 10.0 from Mestrelab.

2D Heteronuclear Overhauser Enhancement Spectroscopy

(32) HOESY of PFCE-particles were measured was measured in D2O as solvent at Bruker DMX 500 MHz NMR spectrometer with relaxation delay D1=2.0 s and cross-relaxation delay D8=0.25 s, 471 MHz (.sup.19F), 500 MHz (.sup.1H)

(33) HOESY of PFCE-loaded core-shell capsules were performed at Bruker Avance III 400 MHz spectrometer with D1=2.0 s and D8=0.15 s, 377 MHz (.sup.19F), 400 MHz (.sup.1H).

(34) Data evaluation was done with MestreNova 10.0 from Mestrelab.

Quantification of PVA

(35) Concentration of PVA was measured with 1 H NMR using resorcinol as an internal reference at Bruker Avance III 400 MHz spectrometer with D2O as solvent. Relaxation delay D1=30 s. Data evaluation was done with MestreNova 10.0 from Mestrelab.

Solid State NMR

(36) Solid state NMR was measured at Varian VNMRS 850 MHz spectrometer at 10 kHz MAS. For measuring the particles in dry state we used freeze-dried nanoparticle powder, which was obtained according to the method described in this patent. To obtain nanoparticles that are swollen with water, freeze-dried powder was incubated with excess of water for 5 min and then centrifuged at 15000 rpm using hettich micro 200 R centrifuge. The pellet from centrifugation was then immediately filled into a rotor for solid state NMR measurements. Data evaluation was performed with MatNMR.

Differential Scanning Calorimetry (DSC)

(37) DSC was measured at Mettler Toledo DSC822e calorimeter equipped with an FRS5 sensor, a Julabo FT900 immersion cooler, a TSO 801RO Sample Robot with heating rate 2 K/min under nitrogen atmosphere using STARe software 11.0 form measurements and data analysis.

Thermogravimetric Analysis (TGA)

(38) TGA was measured at Mettler Toledo TGA/SDTA851e instrument under nitrogen atmosphere with a heating rate of 20 K/min.

Dynamic Light Scattering (DLS)

(39) DLS was performed at zetasizer ZS nano from Malvern instruments at sample concentration of 0.1 mg/mL in deionized water.

Small Angle X-ray Scattering (SAXS)

(40) SAXS measurements were performed at Ganesha X-ray instrument equipped with a GeniX-Cu ultra low divergence source (I=1.54 Å, flux of 1×108 ph/s), a Pilatus 300 K silicon pixel detector (487×619 pixels of 172×172 μm2), Linkam temperature controller (−80-250° C.), Julabo temperature controller (−5-80° C.), q-range 0.003-3 Å.sup.−1.

X-ray Diffraction (XRD)

(41) Diffractograrn was measured on a Panalytical Empyrean in reflection mode with fine-focus sealed tube, and PIXcel3D detector, using CuKα radiation. The scan range was from 2 to 50 degrees 2-theta, with a step size of 0.013 degrees.

Characterization

(42) The beads obtained according to the process of the invention were characterized by various methods.

(43) The particles prepared in exp 13, which is an experiment from WO2014/041150, are comprised of PLGA, PFCE and Gd-chelate. In contrast, to these particles the new beads (see experiments 1-6) contain poly(vinyl alcohol) (PVA) that was used as stabilizer during the emulsification process to facilitate the formation of stable and monodisperse particles.

(44) Thus, the process according to the present invention results in particles with a different composition then the process according to WO2014/041150. This changing of particle composition can be demonstrated by .sup.1H NMR measurements. FIG. 2 shows quantitative .sup.1H NMR spectra of particles form the old patent and the new particles with resorcinol as an internal reference. Due to reduced mobility of PLGA chains after particle formation, PLGA cannot be detected by NMR, and only signals of resorcinol and PVA are visible on both spectra.

(45) 1H NMR spectra of particles produced by the process of WO2014/041150 (left) and new (right) process have been measured (see FIG. 1). The spectra were measured with resorcinol (4.1 mg) as an internal reference; 400 MHz, D.sub.2O, relaxation decay=20 s. Inset in the right spectrum: chemical structure of PVA, non-hydrolyzed OH groups not shown for simplification.

(46) The peak at 4.06 ppm corresponds a proton of H—COH group of a monomer repeat unit, as shown on the inset in FIG. 1 right. Using the integral of the triplet signal of resorcinol at 7.18, the integral of the PVA peak at 4.06 ppm and the average molecular weight of PVA 9500 g/mol (molecular weight by manufacturer 9000-10000 g/mol) can be calculated resulting in following PVA content: particles from the process of WO2014/041150: 7 wt.-% particles from the new process: 23 wt.-%

(47) Solid state NMR measurements and 2D solution NMR that are shown in the description demonstrated that PFCE interacts with water. As PLGA is a hydrophobic polymer that is not soluble in water while PVA is a hydrophilic polymer, we assume that PVA that is present in new particles promoted this interaction between PFCE and water resulting in a unique structure. PFCE is an ultrahydrophobic compound that usually does not mix with water. Thus, particles or emulsions reported until now in the literature typically have a core-shell structure.

Experiment 15 (Comparative)

(48) To prove the unique structure of our particles, we prepared core-shell capsules as an additional control. To make these core-shell capsules, we used the procedure for synthesis PFOB-loaded capsules with sodium cholate as surfactant that was previously described by Pisani et al (Adv. Funct. Mater. 2008, 18, 2963-2971). In this procedure we replaced PFOB by 0.9 mL PFCE, to make PFCE capsules, which can then be directly compared to our PFCE beads.

(49) PLGA (100 mg, resomer 502 H) was dissolved in 3 mL dichloromethane and mixed with perfluoro-15-crown-5 ether (900 μL) or perfluorooctylbromide (PFOB, 275 μL) by pipetting it up and down with a glass pipette. The resulting primary emulsion was added to 1.5 wt.-% solution of sodium cholate and sonicated on ice for 3 min at amplitude of 40% (branson digital sonifier s250). After sonication dichloromethane was evaporated over night under stirring at room temperature. To exchange the surfactant, PVA solution (10 g of 1.96 wt.-%) was added to the suspension and the mixture was stirred at 4° C. for 5 d. The emulsions were washed 2 times (with water at 16000 g, resuspended in 4 mL of water on a shaker at 4° C., frozen with liquid N2 and freeze-dried.

(50) The synthesis yielded freeze-dried capsules, with R.sub.h=62 nm and PDI 0.09.

(51) In contrast to our particles this nano-capsules show almost no HOESY between F-atoms in PFCE and water (FIG. 9) in NMR. Due to core-shell structure of these capsules, water cannot penetrate inside the superhydrophobic PFCE core. Therefore, almost no interaction between water and PFCE is present. The slight residual HOESY signal could result from the contact area of the core with the shell i.e. the interface of the PFCE filling with the PLGA shell. In contrast our particles show extensive HOESY between water protons and fluorine atoms in PFCE (FIG. 3). Due to the homogenous multi-domain structure of our particles that is stabilized by hydrophilic PVA, the contact area between water and PFCE is much higher resulting in stronger HOESY.

Remarks

(52) HOESY cannot be measured with Gd-containing particles, as Gd is paramagnetic. The prior art (Pisani) missed essential details on how to perform the experiments. The inventors used their sonicator for making these capsules, so they could not use the exact sonication settings from literature. They do not believe that this is important for obtaining capsules.

(53) The method the inventors used was based on several papers from that group in combination of with one of their PhD thesis. The problem was that they often published non-completed methods (e.g. skipped volume of surfactant).

(54) ssNMR of swollen capsules could not be measured, as they are not that stable in concentrated aqueous solution and ssNMR takes several hours.