ULTRA-SMALL SUPERPARAMAGNETIC IRON OXIDE NANOPARTICLES

Abstract

The invention relates to a method for producing an adjusted nanoparticle composition comprising ultra-small superparamagnetic iron oxide nanoparticles coated with dextran-T10, compositions obtained thereby, and uses of such compositions. The adjusted compositions have improved parameters such as lower batch-to-batch variance and improved stability, and are useful as magnetic imaging agents.

Claims

1. Method for producing an adjusted nanoparticle composition, the method comprising the steps of: i) providing an ultrafiltrated composition comprising ultra-small superparamagnetic iron oxide nanoparticles coated with dextran-T10; ii) determining the amount of dextran-T10 present in the ultrafiltrated composition of step i); iii) adding an amount of dextran-T10 to the ultrafiltrated composition of step i) to obtain an adjusted composition comprising at least 140 and at most 160 wt.-% of dextran-T10 relative to the wt.-% of iron; and iv) optionally filtering the adjusted composition.

2. The method according to claim 1, wherein the ultrafiltrated composition of step i) or the adjusted composition of step iii) comprises from 1.5-2.5 wt.-% iron.

3. The method according to claim 1, wherein step ii) further comprises determining the amount of iron in the ultrafiltrated composition of step i).

4. The method according to claim 1, wherein step iii) further comprises adding a tonicity agent such as a citrate to the ultrafiltrated composition or to the adjusted composition, preferably in an amount of 7-12 wt.-% relative to the wt.-% of iron.

5. The method according to claim 1, wherein the ultrafiltrated composition of step i) further comprises a pharmaceutically acceptable excipient.

6. The method according to claim 1, wherein step iii) further comprises adding an amount of dextran-T1 to the ultrafiltrated composition or to the adjusted composition, so that the adjusted composition comprises at least 140 and at most 160 wt.-% dextran-T1 relative to the wt.-% of iron.

7. The method according to claim 6, wherein the adjusted composition comprises a substantially equal amount by weight of dextran-T10 and of dextran-T1.

8. The method according to claim 1, wherein in step i) the ultrafiltrated composition is provided by a method comprising the steps of: i.a) providing a solution of dextran-T10, FeCl.sub.3, and FeCl.sub.2 in water; i.b) adding a base such as ammonium hydroxide to the solution of step i.a); i.c) heating the solution of step i.b) to over 60° C. to obtain a nanoparticle composition comprising ultra-small superparamagnetic iron oxide nanoparticles coated with dextran-T10; i.d) purifying the nanoparticle composition of step i.c) using ultrafiltration to obtain an ultrafiltrated composition.

9. The method according to claim 1, wherein a filtrate of the ultrafiltrated composition of step i) has a conductivity of less than 500 μS/cm, preferably of less than 50 μS/cm, more preferably of less than 25 μS/cm.

10. The method according to claim 1, further comprising the step of lyophilizing the adjusted composition.

11. Composition comprising: i) ultra-small superparamagnetic iron oxide nanoparticles coated with dextran-T10; ii) free dextran-T10; iii) optionally a tonicity agent such as a citrate; wherein the composition comprises at least 140 and at most 160 wt.-% of dextran-T10 relative to the wt.-% of iron.

12. (canceled)

13. The composition according to claim 11, wherein the concentration of dextran-T10 has a dispersion of at most ±10%.

14. The composition according to claim 11, wherein the unimodal mean diameter of the nanoparticles is stable for at least 6 months.

15. (canceled)

16. A method of performing an MRI, comprising administering a contrast agent to a subject, wherein the contrast agent is the composition according to claim 11.

Description

DESCRIPTION OF DRAWINGS

[0129] FIG. 1—Sketch of an iron oxide core stabilised by dextran. Parts are not to scale.

[0130] FIG. 2—Flowchart depicting a method according to the invention. The steps with a bold dashed outline are the steps wherein an adjusted nanoparticle composition is produced.

[0131] FIG. 3—Analysis of different dextran fractions after ultracentrifugation of an adjusted nanoparticle composition. Dextran-T1 is revealed to remain substantially unassociated with the nanoparticles, whereas about 70% of Dextran-T10 is associated with the nanoparticles.

[0132] FIG. 4—Analysis of the influence of Dextran-T1 or of citrate on different parameters of the adjusted nanoparticle composition. A) influence of various concentrations of Dextran-T1 and citrate on pH; no clear optimum can be determined. B) influence of various concentrations of Dextran-T1 and citrate on unimodal diameter; an optimal zone can be found when Dextran-T1 is at 20 to 30 mg/g while citrate is beneath 2.6 mg/g. C) influence of various concentrations of Dextran-T1 and citrate on nanoparticle-bound citrate levels; dextran-T1 has no significant effect on this parameter, and citrate in excess of about 2.1 mg/g has no more significant effect.

EXAMPLES

Example 1—Production of Adjusted dUSPIO

Provision of Ultrafiltrated dUSPIO (First Seven Shapes in FIG. 2)

[0133] The description of the manufacturing process given hereafter is for a clinical batch size (143 g iron).

[0134] Step 1: Preparation of Dextran-Iron solution—Purified water (1950±40 g) is added into a 5 L container labeled A. While mixing the purified water, Dextran-T10 (1143 g) is added to the container. Ferric chloride (84.7±1.7 g) is added to the Dextran solution and the solution is mixed for a minimum of 3 hours. Transfer the Dextran-Iron solution to container “B” through a 0.2μ filter and a 0.45 μm prefilter, then transfer into the reactor using a peristaltic pump. The Dextran-Iron solution is cooled under nitrogen with stirring in the reactor until the temperature reaches 1-6° C.

[0135] Step 2: Preparation of Ferrous Chloride Solution—Purified water (46.8±0.9 g) is added into the 250 mL container labelled C. The purified water is stirred while 34.8 g±0.7 g of ferrous chloride tetrahydrate is added. The stirring is continued for a minimum of 110 minutes. The solution is stored under nitrogen at 2-8° C. until needed. Step 3: Preparation of Strong Ammonia solution—Strong ammonia solution (103.2±2.0 g) is weighed in the fume hood into a 250 mL container labelled D and cooled to 2-8° C.

[0136] Step 4: Heating to 80° C.±5° C.—The reactor contents are stirred under nitrogen while the ferrous chloride solution is pumped through a 0.2μ filter into the reactor. The ammonium hydroxide solution is then added to the reactor. The reactor is heated and stirred until the contents reach 80° C.±5° C.

[0137] Step 5: Dilution with purified water—8660±150 mL of 80° C.±5° C. purified water is then added to the reactor with continued stirring. The temperature of the reactor is cooled to 25° C.±5° C. The solution is filtered via a 0.2 μm filter (and a 0.45 μm prefilter) into container F.

[0138] Step 6: Ultrafiltration—Six superparamagnetic iron oxide batches are filtered through a 0.2μ filter (and a 0.45 μm prefilter) directly into the tank of the ultrafilter apparatus fitted with a 100,000 Dalton cut-off cartridge membrane. A constant flow ultrafiltration procedure is used for purification. The solution is processed through the ultrafilter until the retained volume is 25.5 L. The effluent is discarded during all the ultrafiltration steps. The iron oxide solution is processed through the ultrafilter again by washing the solution with 51 L water for injection while keeping the volume at 25.5 L (continuous ultrafiltration). The solution is processed through the ultrafilter until the retained volume is 9 L. The iron oxide solution is processed through the ultrafilter again by washing the solution with 24 L water for injection while keeping the volume at 9 L. A minimum of two more cycles of ultrafiltration is performed. The conductivity of the effluent is tested. If it is less than 25 μS, the ultrafiltration is continued to a volume of 2.16±0.1 L and the ultrafiltration is complete. If the conductivity is greater than 25 μS, either one or two additional ultrafiltration cycles are performed with conductivity testing after each cycle. Both conditions must be satisfied for ultrafiltration to be complete: [0139] First, a minimum of 10 diafiltration volumes of water for injection must have been used, [0140] Second, a conductivity of less than 25 μS must be reached. The solution is removed from the ultrafilter tank and filtered via a 0.2 μm filter (with 0.45 μm prefilter) into a 10 L container. Water for injection (1440±40 mL) is added into the ultrafilter tank to rinse it and then added to the 10 L container.

Analysis of Ultrafiltrated dUSPIO (Shape 8 in FIG. 2)

[0141] Step 7—The ultrafiltrated solution is stirred for a minimum of 5 minutes and then analysed for total iron concentration and dextran content using Atomic Adsorption Spectroscopy for determination of iron content and TOC (total organic carbon) for total dextran content. The solution is stored at 2° C.-8° C. until the results are available.

Adjustment of dUSPIO and Further Processing (Shapes 9, 10, part of 11 in FIG. 2)

[0142] Step 8: Preparation of Dextran-T10, Dextran-T1 and sodium citrate dihydrate solution—The amount of a Dextran-T10, Dextran-T1 and sodium citrate dihydrate solution in water for injection is calculated based on the iron concentration and dextran-T10 concentration from step 7 such that in the final formulation the concentration of iron is 18-20 mg/g, the total concentration of dextran is 53-61 mg/g, the ratio of dextran-T10 and dextran-T1 is 1/1 g/g and the citrate content is 1.5-2.1 mg/g. Water for injection is added into a 10 L container. The water for injection is stirred, and Dextran-T10, Dextran-T1, and sodium citrate dihydrate are added (good results were obtained with slow addition). The solution is stirred for a minimum of 15 minutes.

[0143] Step 9: Addition of Dextran-T10, Dextran-T1 and sodium citrate dihydrate Solution—The solution of step 7 is mixed and the final formulation is prepared by adding the solution of step 8 through a 0.2 μm filter and a 0.45 μm prefilter while continuously mixing.

[0144] Step 10: filtration—The solution is filtered through a 0.2μ filter into 10 L containers for storage and shipping. This step can be repeated if necessary (e.g. in case of filter failure).

[0145] Table 1 shows examples of adjustment of three different batches that were prepared according to steps 1-6 above.

TABLE-US-00001 TABLE 1 examples of the preparation of adjusted batches Action Batch 1 Batch 2 Batch 3 Conductivity final ultrafiltration 6 17 9 flowthrough (μS/cm) Total weight of batch (g) 2542 3582 3730 Fe concentration (mg/g) 41.0 42 40.5 Dextran-T10 content (mg/g) 31 33 31 Total Dextran-T10 after 78.8 118.2 115.6 ultrafiltration (g) Dextran-T10 added for 77.7 107.4 110.9 adjustment (g) These compositions were further treated as follows: Dextran-T1 added (g) 156.1 225.5 226.6

Example 2—Comparison of Adjusted dUSPIO to Unadjusted dUSPIO

[0146] The adjusted composition of example 1 is an optimised formula because it is adjustment for dextran-T10 during the manufacture. This adjustment allows to ensure a better reproducibility of the concentration of dextran in the finished product (±10%) and increases stability of the unimodal mean diameter. To demonstrate that the formulae are otherwise equivalent, a comparison study was performed between adjusted (entire protocol of example 1) and unadjusted (steps 1-6 of example 1) compositions. Both compositions comprised 1.8 mg/g citrate ion and 28.5 mg of dextran-T1 and were stabilised as a lyophilizate and stored at 4° C. and 25° C. for 6 months. Zeta potential was measured using a Malvern Zetasizer. Unimodal mean diameter was measured using Laser Light Scattering (suitable models are Malvern Zetasizer, Brookhaven BI 90, and Malvern 4700).

TABLE-US-00002 TABLE 2 comparison of adjusted and unadjusted dUSPIO Unadjusted Adjusted 0 1 3 6 Months 0 1 3 6 8.2 8.2 8.2 8.1  pH 4° C. 8.0 8.1 8.1 8.1 8.2 8.1 8.1 8.1 pH 25° C. 8.0 8.1 8.1 8.1 36 36 39 37 Unimodal mean 36 36 36 38 diameter 4° C. 36 36 40 25 Unimodal mean 36 36 38 36 diameter 25° C. −47.4 Zeta potential −51.4 water −19.9 Zeta potential −21.3 0.9% NaCl

[0147] For both compositions, at 4 and 25° C., the pH does not vary in time, and the zeta potential is highly similar in both water or salt solution. For the unadjusted composition the unimodal mean diameter varied more than 7% at 4° C., and more than 35% at 25° C., while under both conditions the adjusted composition remained at about 5% variation.

[0148] Unadjusted and adjusted batches were also analysed for their total dextran content. Adjusted compositions could be specified at ±10% having a dispersion of ±9%. Unadjusted compositions had a dispersion of ±17%.

TABLE-US-00003 TABLE 3 comparison of dextran concentrations Unadjusted Adjusted Total dextran Total dextran (mg/g) Batch No. (mg/g) 50 1 56 51 2 54 53 3 54 49 4 54 48 5 58 45 6 55 49 Mean 55 8.2 3x σ 4.8 ±17% Dispersion ±9%

Example 3—Analysis of dUSPIO

Analysis of Free and Associated Dextran Fractions

[0149] Dextran fractions of the composition as prepared in example 1 were determined after a single step of ultracentrifugation at 30 kDa MWCO as described above for the ultrafiltrated composition, after having allowed the composition time to equilibrate. Dextran was identified using gel permeation chromatography, results are shown in FIG. 3. Dextran-T1 is revealed to remain substantially unassociated with the nanoparticles, whereas about 70% of Dextran-T10 is associated with the nanoparticles. Dextran-T1 remains at the free state in the solvent and does not alter the dextran-T10/iron oxide interaction. The data also confirm that the particle-bound fraction of dextran-T10 represents 70% of the total quantity of dextran-T10 when the composition is in equilibrium.

Determination of Optimal Excipient Concentrations

[0150] Different formulations were prepared as shown in Table 4, to be used for analysis according to a Doelhert's matrix (see for example Sautour et al., J. App. Microbiol., 2001, 91, 900-906). Each formula was stabilised as a lyophilizate, and was analysed after reconstitution in 0.9% NaCl. Each formula was stored at room temperature and at 55° C. for 12 months. The following parameters were analysed: pH, particle size by laser scattering, zeta potential, and bound citrate. Bound citrate can be assessed by HPLC techniques, or LCMS, on supernatant after reconstitution. The results are analysed using Nemrod 3.0 software.

[0151] The factors (concentration in citrate and concentration in dextran-T1) are symbolised by X1 et X2, respectively. The function that models the Y responses (parameters) measured in terms of X1 and X2 is:

Y=b.sub.0+b.sub.1X.sub.1+b.sub.2X.sub.2+b.sub.11X.sub.1.sup.2+b.sub.22X.sub.2.sup.2+b.sub.12X.sub.1X.sub.2

TABLE-US-00004 TABLE 4 compositions as used in this example Composition Unimodal diameter Bound citrate Citrate Dextran-T1 pH T = 0 25° C. 55° C. 25° C. (mg/g) (mg/g) T = 0 25° C. 55° C. (nm) (nm) (nm) (mg/g) 0.77 14.4 7.87 7.88 6.99 36  46  59 0.69 1.22 1.9 7.99 8.05 7.60 53 101 157 0.91 1.22 26.8 8.07 8.03 7.01 30  30  31 0.90 1.67 14.4 8.09 7.97 7.37 32  25  38 1.09 1.67 14.4 8.10 7.98 7.38 32  34  38 1.08 2.13 1.9 8.15 8.21 7.56 42  68 107 1.16 2.13 26.8 8.10 8.10 7.35 32  32  34 1.12 2.58 14.4 8.19 8.11 7.61 34  37  41 1.19 1.81 28.7 8.08 8.04 7.35 28  30  31 1.12

pH

[0152] In any experiment performed, no variation of the pH at 25° C. is observed. This parameter is not relevant in this study. The response of pH (55° C.) is modelled by the following polynomial:

pH=7.38+0.26X.sub.1−0.23X.sub.2−0.08X.sub.1.sup.2+0.03X.sub.2.sup.2+0.22X.sub.1X.sub.2

[0153] The equation shows that the main effects are related to citrate (increased pH when the concentration increases), to dextran-T1 (decrease when the concentration increases) and to the interaction of both factors. The response curves (FIG. 4A) illustrate these results, without showing any characteristic optimum. At high temperature, citrate acts as a buffer, allowing limitation of the decrease in pH related to the presence of dextran. This phenomenon is not visible in normal conditions for storage.

Unimodal Diameter

[0154] The response of the mean unimodal diameter (25° C.) is modelled by the following polynomial:

Diameter=34.1−8.3X.sub.1−30.5X.sub.2+7.0X.sub.1.sup.2+29.2X.sub.2.sup.2+20.1X.sub.1X.sub.2

[0155] The equation shows that dextran has the major effect. There is a slight interaction between citrate and dextran-T1. The response curves (FIGS. 4B) clearly show an optimal zone for dextran-T1 contents from 20 to 30 mg/g and for citrate contents inferior to 2.6 mg/g.

[0156] The response of mean unimodal diameter (55° C.) is modelled by the following polynomial :

Diameter=37.9−13.8X.sub.1−57.8X.sub.2+12.1X.sub.1.sup.2+55.2X.sub.2.sup.2+30.5X.sub.1X.sub.2

[0157] The effects are more marked than at 25° C., but their nature is unchanged.

Zeta Potential

[0158] The response of potential is modelled by the following polynomial:

ζ=−42.5−6.1X.sub.1−3.3X.sub.2+11.7X.sub.1.sup.2−0.5X.sub.2.sup.2+3.1X.sub.1X.sub.2

[0159] The equation shows that citrate has the major effect. The optimum (maximum load) is within a range comprised between 1.4 and 2.3 mg/g, concentrations above 2.3 mg/g (12.12 wt.-% vs iron) have no further effect.

Bound Citrate

[0160] The response of bound citrate is modelled by the following polynomial:

Bound citrate=1.09+0.25X.sub.1−0.01X.sub.2−0.15X.sub.1.sup.2−0.04X.sub.2.sup.2−0.03X.sub.1X.sub.2

[0161] As expected, the equation shows that only citrate has an effect on this parameter. The response curves (FIG. 4C) also show a phenomenon of saturation in citrate for the particle surface. The optimal response is from 0 to 1.1 mg/g, and it is obtained from concentrations in total citrate equal to 1.4 mg/g. Above 2.1-2.2 mg/g the excess total citrate has no more significant effect. These results are consistent with those obtained for the zeta potential.

Conclusion

[0162] The study showed that citrate contributes to particle charge and it is beneficial at a concentration superior to 1.4 mg/g; dextran-T1 acts as a cryoprotector during lyophilization and is most efficient at a concentration superior or equal to 20 mg/g.

Example 4—Lyophilisation of dUSPIO

[0163] Three different compositions were analysed for their thermal properties using standard techniques for monitoring ice matrix changes by examining electrical conduction properties. The composition comprising both dextran-T1 and citrate was found to be more resistant to lyophilisation, beyond what could be expected based on the additive effect of the individual additives. This translates into a slower and more gradual lyophilisation, which reduces particle aggregation.

TABLE-US-00005 TABLE 5 thermic properties of compositions as described herein Dextran + Parameter Dextran Citrate citrate Iron content 20 mg/mL 20 mg/mL 20 mg/mL Dextran-T1 20 mg/mL — 20 mg/mL Citrate — 25 mM 25 mM Freezing temperature −50° C. −35° C. −50° C. Collapse temperature −13° C. −20° C. −43° C. Incipient melting −4° C. −10° C. −9° C. temperature Sublimation −18° C. −25° C. −48° C. temperature Recommended 200 mTorr 200 mTorr 20 mTorr pressure for lyophilisation Recommended 12 h 10 h 20 + h duration of lyophilisation

TABLE-US-00006 References EP0713602B1 U.S. Pat. No. WO 97/25073 WO 96/09840 5,262,176 [0164] Anzai et al., 1994, AJNR Am J Neuroradiol 15:87-94 [0165] Dubois et al. (1956, Anal. Chem., 28:350-356) [0166] Corot et al., doi:10.1016/j.addr.2006.09.013 [0167] Laurent et al., doi: 10.1021/cr068445e [0168] Jung, 1995, Magnetic Resonance Imaging, Vol. 13, No. 5, pp. 675-691 [0169] Jung and Jacobs, 1995, Magnetic Resonance Imaging, Vol. 13, No. 5, pp. 661-674 [0170] Merup & Hansen, 2007, “Superparamagnetic particles”, doi: 10.1002/9780470022184.hmm409 [0171] Paul et al., doi: 10.1021/bc034194u Saleh et al., doi:10.1002/nbm.881 [0172] Sautour et al., J. App. Microbiol., 2001, 91, 900-906 [0173] Sigal et al., doi: 10.1007/s003300101130 Weinstein et al., doi:10.1038/jcbfm.2009.192 [0174] Unterweger et al., doi: 10.2147/IJN.S138108