WO2009156445A2

WO2009156445A2 - Use of cobalt ferrites as contrast agents for magnetic resonance

Info

Publication number: WO2009156445A2
Application number: PCT/EP2009/057908
Authority: WO
Inventors: Giovanni Baldi; Daniele Bonacchi; Marco Bitossi; Alessandro Lascialfari
Original assignee: Colorobbia Italia S.P.A.
Priority date: 2008-06-26
Filing date: 2009-06-24
Publication date: 2009-12-30
Also published as: WO2009156445A3; ITFI20080117A1

Abstract

There are described contrast media for magnetic resonance comprising nanometer-sized cobalt ferrites, their preparation and use.

Description

USE OF COBALT FERRITES AS CONTRAST AGENTS FOR MAGNETIC RESONANCE

Field of the invention The present invention relates to the field of contrast agents for magnetic resonance. State of the art

As is known, the MRI technique enables images of inert materials and/or living beings to be generated utilizing the density of protons (hydrogen nuclei) or other nuclei by using magnetic field gradients and suitable radio frequency pulse sequences. In particular the contrast generated by different parts of the material, characterized by different spin-spin (T₂) and spin-lattice (T₁) nuclear relaxation times, can be utilized to better display the images and to gain specific information. For the same experimental parameters, the "natural" imagine contrast is enhanced by the use of contrast agents (or media) (CA). Contrast agents increase the contrast between normal and damaged tissues and generally shorten the T₁ and T₂ nuclear relaxation times of the liquids contained within tissues. The parameters which characterize contrast agents and measure their efficiency are transverse (R₂) and longitudinal (R₁) nuclear relaxivity, defined as the increase in solvent (water) proton relaxation rate produced by 1 mmol/l of paramagnetic centre.

The ideal materials for achieving faster nuclear relaxation are structures rich in unpaired electrons i.e. paramagnetic substances. Said substances, characterized by a magnetic centre and an organic coating, are generally CA positive. Other contrast agents based on superparamagnetic particles (generally ferrites) exhibit high magnetic moments and are generally CA negative. Currently, the majority of paramagnetic compounds in use are gadolinium based (Gd-DTPA, Gd-DTPA-BMA, Gd-HP-DO3A), the component molecules having mean diameters of about 0.35 nm. The most widely used superparamagnetic particles in resonance experiments are known commercially as Endorem® (by Guerbet Group, Feridex® in the USA), Resovist® (by Schering) and Sinerem® (by Guerbet Group, Combidex® in the USA); in each of these the central ferrite core (diameter about 5-6 nm) is coated with an organic shell, the mean final diameters being respectively -150 nm and -15 nm. From this characteristic mean diameter are derived the particle names SPIO (superparamagnetic iron oxide particles) for Endorem® and Resovist®, and USPIO (ultrasmall superparamagnetic iron oxide particles) for Sinerem®. To increase the stability of these particles in aqueous solution, they are coated with a polymer (dextran) to form colloidal solutions. After intravenous injection, these particles are trapped by reticuloendothelial cells (such as Kupffer cells in the liver, and in the spleen). They act as negative contrast agents and induce a signal reduction (dark areas) in these organs; they are generally used for detecting hepatic tumours. The tendency of current research is to establish a connection between these materials and specific molecules able to target specific cell receptors that are characteristic of a particular pathology. MRI applications of these superparamagnetic compounds are varied, ranging from angiography to the diagnosis of tumours and atherosclerotic pathologies. Progress in molecular biology has led to new and specific diagnostic imaging methods, of which Molecular Imaging is one of the most promising due to the use of specific labels and its excellent spatial resolution. Recent research on cell labelling in MRI has made use of a compound which targets specific cell receptors by way of ligands chemically bound to SPIO or USPIO particles. Moreover, it is known that magnetic particles can absorb electromagnetic radiation and cause local heating (magnetic hyperthermia) [Rosenweig, R. E. J. Magn. Magn. Mater. 2002, 252, 370]. This effect can be utilized for cancer therapy both for the hyperthermic effect itself (temperature-induced apoptotic or necrotic effect) and as a means for controlling localized drug delivery. In the light of the aforesaid, it is obviously important to be able to provide ever more effective contrast media to better meet the need for a clear picture of the organs to be examined by the aforesaid techniques, and in addition to provide contrast media able to also simultaneously exert a hyperthermic effect and hence able to simultaneously perform the function of contrast agent and therapeutic (theranostic) compound.

Brief description of the figures

Fig. 1 shows the transverse relaxivity of two products of the invention (E7Vvac and NBR1 ).

Fig. 2 (a-c) gives a comparison of the MRI signal obtained with GE-PDW sequences, between NBR1 and the commercial compound Endorem at different concentrations. Fig. 3 (a-d) shows images of the brain (left) and liver (right) respectively: immediately after injection with NBR1 (a), after 1 day (b), after 6 days (c) and after 6 days with Endorem use (d). Detailed description of the invention The present invention therefore relates to contrast agents having a particularly rapid nuclear relaxation, also usable in hyperthermia therapy.

It has been surprisingly found that magnetic cobalt ferrite nanoparticles of mean diameter between 4 and 100 nm have a transverse relaxivity R2 value, and a R2 to R1 ratio greater than that of the iron oxide-based contrast agents currently available on the market (Endorem). Moreover, these properties remain unchanged even when said cobalt ferrite nanoparticles are functionalized and/or incorporated in constructs of composite materials.

The most common method for producing magnetic nanoparticles involves the co- precipitation of ferrous and ferric salts in an alkaline medium in the absence or presence of surfactants such as PEG, dextran, synthetic polymers, etc. In methods not using surfactants, the precipitated nanoparticles are isolated by magnetic separation or centrifugation. The precipitate is then treated with nitric acid, centrifuged and peptidized in water to produce a stable acid magnetic sol. In the synthesis of iron oxide particles many parameters can be adjusted to control their dimensions, magnetic characteristics, stability in solution and surface properties. Although co-precipitation methods are used for their simplicity, the nanoparticles produced prove to be polydisperse.

A further preparation method, which enables the aforesaid problem to be eliminated, is the polyol preparation method as described for example in patent application PCT/EP2007/064143 in the name of the same applicant and briefly summarized herein for completion. This method in fact enables magnetic nanoparticle dispersions to be synthesized with a narrow size distribution and hence high contrast efficacy. The magnetic nanoparticles according to the invention may be prepared according to known processes such as the polyol process widely described in the literature which, briefly, consists in the use of a high boiling alcohol allowing operation at high temperatures and complexing of particles during their formation, thus impeding their growth.

Normally, the desired metal precursors (preferably acetates, carbonates, sulphates, oxalates, chlorides) are added to a known volume of alcohol (e.g. diethylene glycol, DEG). The solution is then heated under agitation until complete solubilisation of the precursors, water is possibly added in an appropriate amount to facilitate hydrolysis of the precursors, followed by heating for a few hours at a temperature higher than 150^QC then allowed to cool, thus obtaining a stable suspension of monodisperse nanoparticles with a narrow size distribution. According to the invention, synthesis methods were established capable of enabling dimensional (and hence hyperthermia) control of the magnetic nanoparticles, which are always obtained in suspension. This can be carried out by using a continuous process.

In this case the procedure is as aforedescribed for the polyol process, but the synthesis is performed by the addition (in a quantity equimolar to the reagents) of seeds consisting of the previously synthesized magnetic nanoparticles. In this manner, at the end of the reaction magnetic nanoparticles greater in size than those introduced at the beginning of the synthesis are obtained. In practice, an initial preparation is carried out as for the polyol process; subsequently, a new reaction is carried out under the same conditions as the first one, with all of the starting materials in amounts identical to those already used and with the addition of the product obtained from the first reaction. The magnetic nanoparticles thus obtained (which are double in quantity and greater in size than those introduced at the beginning of synthesis) can be used again as seeds in the subsequent reaction. The cycle can be repeated indefinitely until particles of the desired size are obtained.

According to a further embodiment of the invention the nanoparticles can be prepared by a process of semicontinuous substitutions. In practice, a first synthesis according to the polyol process is performed, but at the end of the stationary heating period at 180^QC the product is not cooled but rather poured into a flask of twice the size into which all of the starting materials have been fed in amounts identical to the already reacted product. The temperature is again brought to 180^QC and maintained for 3 hours, then the cycle is repeated for a variable number of times until obtainment of the desired product dimensions.

A further preparation process of the particles of the invention is a growth process.

In this case the synthesis is conducted according to the aforedescribed polyol process but the period during which the product is maintained at a temperature of

180^QC is extended for a variable number of hours. A product is thus obtained the dimensions of which are dependent on the residence time at that temperature.

The magnetic nanoparticles can also be prepared by a process similar to the above described polyol process, though performing the heating exclusively in a microwave oven which allows the reaction times to be considerably reduced and to have a better control of the size and morphology.

Functionalization of the nanopartides, when desired or necessary, was obtained according to known methods i.e. by reacting monofunctional or difunctional derivatives with the nanopartides as defined above so as to coat their surface as described in patent application PCT/EP2007/050036 in the name of the same applicant.

Examples of mono- and difunctional compounds are compounds of general formula:

in which: n is an integer between 2 and 20;

R₁ is chosen from: CONHOH, CONHOR, PO(OH)₂, PO(OH)(OR), COOH, COOR,

SH, SR;

R₂ is the external group and chosen from: H, OH, NH₂, COOH, COOR; R is a C1 -6 alkyl group such as ethyl, or an alkali metal, preferably K, Na or Li.

As stated in the aforementioned patent application the derivatization takes place by reacting a nanoparticle dispersion in an organic solvent (such as ethylene glycol) with the chosen ligand while stirring under reduced temperature for a few hours. The product is then possibly separated by extraction with specific solvents or precipitated, for example with acetone, then centrifuged, separated and possibly redispersed in a suitable solvent. The previously mentioned polymer constructs have different characteristics depending on the type of polymer used for their preparation as described in the aforesaid patent application PCT/EP2007/064143.

They consist of magnetic nanoparticles, possibly functionalized, incorporated in a water soluble or water insoluble polymer. Water soluble polymers according to the invention can be surfactants, polyelectrolytes, water soluble polypeptides and proteins, preferred are water soluble polymers chosen from block copolymers, modified polyethylene glycols, modified polysaccharides, phospholipids, polyaminoamides, globular proteins. Usable water insoluble polymers according to the invention include polyesters, polyamides, polyanhydrides, polyorthoesters, peptides, polyaminoamides, preferred are polymers chosen from polyesters and polyaminoamides; said insoluble polymers can themselves be stabilized with surface active agents such as: surfactants, polyelectrolytes, polypeptides and water soluble proteins; preferred are surface active agents chosen from block copolymers, modified polyethylene glycols, modified polysaccharides, phospholipids, polyaminoamides, globular proteins.

The Co-ferrite particles of this invention show superparamagnetic properties and hence fall within the class of superparamagnetic agents tendentially of negative contrast action. As already noted above, in order to determine the efficiency of said CAs, the relaxometric efficiency must be evaluated by means of the transverse relaxivity value R2 and the R2/R1 ratio. For this purpose we have used the commercial CA Endorem® as reference material with which to compare the relaxivity values obtained in the newly synthesized compounds. Said material was also used as a reference in the production of phantom images using a low field MRI scanner and of animal models by a high field scanner.

The experiments undertaken have employed: (a) the relaxometry technique by Fourier transform NMR spectrometry called Apollo-Tecmag, Spinmaster-Stelar or MSL200-Bruker with variable field (i.e. frequency) electromagnet or superconductor of about 0.2 to about 70 Mhz; (b) the MRI technique on phantoms using a low-field 0.2 Tesla ArtoscanΘ-Esaote Imager; (c) the MRI technique on animal models using a 4.7 Tesla Avance200-Bruker Imager. The experiments were conducted on samples of: cobalt ferrite nanoparticles (referred to henceforth as c4b9) of crystalline diameter

7.2 nm, dispersed in glycols (Example 1 ); cobalt ferrite nanoparticles c4b9 incorporated in polymers (referred to as E7Vvac) with mean hydrodynamic diameter of 145 nm, dispersed in water; cobalt ferrite nanoparticles of crystalline diameter 6.7 nm, incorporated in biocompatible polymers (referred to as NBR1_32C C2) of average hydrodynamic diameter 93.40 nm (Example 3) dispersed in water.

Example 1

Preparation of cobalt ferrite by the method of semicontinuous substitutions Product: c4b1

Stage 1 (Product: c4b1 )

Reagents:

Fe: Co ratio of 2:1

9.53 g Co(Ac)₂.4H₂0 (23.7% Co w/w) Co(II) = 2.259 g = 0.038 moles

21.42 g Fe(CH₃COO)₃ (Shepherd paste; c. 20% Fe w/w)

Fe(III) = 4.284 g = 0.077 moles

269.04 g DEG

Synthesis: Stage 1

A 4-neck flask is equipped with a stirrer, a bulb condenser equipped with valve for possible distillation, a probe and a stopper (additional neck).

The reagents are poured into the flask with DEG and the mixture brought to 1 10^QC until solubilised (1 hour). The temperature is then increased to 180^QC maintaining at reflux for 3 hours under agitation until a black suspension is formed.

Stage 2 (Product: c4b2) Reagents:

Fe:Co ratio = 2:1

19.06 g Co(Ac)₂-4H₂0 (23.7% Co w/w)

Co(II) = 4.518 g = 0.076 moles 42.84 g Fe(CH₃COO)₃ (Shepherd paste; c. 20% Fe w/w)

Fe(III) = 8.568 g = 0.154 moles

538 g DEG

575 g c4b1

Synthesis: A 2 litre 4-neck flask is equipped with a stirrer, a bulb condenser equipped with valve for possible distillation, a probe and a stopper (additional neck). The cobalt acetate and ferric acetate are placed into the flask with DEG and the c4b1 product is added, while still hot from the first stage.

The temperature is brought to 180^QC and maintained at reflux for 3 hours. 1 105 g of product are obtained.

Stage 2 is repeated seven times and the final product is named c4b9.

Example 2

Functionalization of cobalt ferrite particles with ethyl-12-(hydroxyamino)-12- oxododecanoate. Product: CoFe38M1

Reagents: ethyl 12-(hydroxyamino)-12-oxododecanoate MW = 273.37g/moles

CoFe₂O₄ (c4b9) MW = 234.62 g/moles

Butanol MW = 74.12 water

Reagents:

60 g of c4b9 in DEG (3% CoFe2O4 w/w) 7.67 mmoles

0.90 g ethyl 12-(hydroxyamino)-12-oxododecanoate 3.29 mmoles

120 g butanol Synthesis:

120 g of butanol and 0.60 g of ethyl 12-(hydroxyamino)-12-oxododecanoate are placed in a 500 ml flask (the mixture completely dissolves in a few minutes). 60 g of a cobalt ferrite nanoparticle dispersion in glycol are added to the solution and the entire mixture is kept under agitation for 2 hours.

The product is washed with 200 g of water (formation of a double phase of butanol/water-glycol) and the alcoholic phase is separated from the aqueous phase using a separating funnel.

The solid product is collected and redispersed in acetone, the butanol being removed under vacuum.

Example 3

Preparation of a construct consisting of nanometer-sized cobalt ferrite particles, PLGA and albumin.

Product: E7Vvac

Reagents: Quantity Molecular weight

UP Water 1000 ml 18 d = 1.00 g/cm³

Acetone 25 ml 58.08 d = 0.79 g/cm³ PLGA 75/25 0.05 g

CoFe38M1 0.02 g

BSA Fraction V 1 g

Synthesis:

A solution of PLGA in acetone (0.05 g in 25 ml of acetone) and a solution of BSA in ultrapure water (1 g of BSA in 1000 ml of water) are prepared, then 0.4 ml of a

5% (w/v) suspension of CoFe38M1 in acetone are added to the PLGA solution.

A double peristaltic pump is installed for continuous addition of acetone solution

(containing PLGA and CoFe38M1 ) to a water flow containing BSA (volume ratio of acetone/water = 1/40). The corresponding immersion tubes draw the solution directly from the tanks containing the two solutions.

The pumping ratio of the two peristaltic pumps is 1/40 such that the two solutions are consumed in the same time period.

The product of the final mixing is collected in a graduated cylinder. The pumping ratio is set so that mixing of the two solutions is achieved in 10 minutes.

The resulting final solution is treated under vacuum so as to completely remove the acetone.

The resulting final solution is concentrated under high vacuum at T < 45^QC or by ultrafiltration until the desired concentration is attained.

Size characterization by means of DLS:

Example 4

Functionalization of cobalt ferrite particles with ethyl 12-(hydroxyamino)-12- oxododecanoate

Product CoFe38H

Reagents: ethyl-12-(hydroxyamino)-12-oxododecanoate MW = 273.37g/moles

CoFe₂O₄ (c4b6) MW = 234.62 g/moles

Butanol MW = 74.12

Water

Reagents:

60 g of c4b6 in DEG (3% CoFe2O4 w/w) 7.67 mmoles

0.90 g ethyl-12-(hydroxyamino)-12-oxododecanoate 3.29 mmoles

120 g butanol

Synthesis:

120 g of butanol and 0.60 g of ethyl-12-(hydroxyamino)-12-oxododecanoate are placed in a 500 ml flask (the mixture completely dissolves in a few minutes).

60 g of a cobalt ferrite nanoparticle dispersion in glycol are added to the solution and the entire mixture is kept under agitation for 2 hours.

The product is washed with 200 g of water (formation of a double phase of butanol/water-glycol) and the alcoholic phase is separated from the aqueous phase using a separating funnel. The solid product is collected and redispersed in acetone, the butanol being removed under vacuum.

Example 5

Preparation of a construct consisting of: nanometer-sized cobalt ferrite, PLGA and albumin

Product : NBR1_32C C2

Reagents Quantity Molecular weight

UP water 1000 ml 18 d = 1 .00 g/cm³

Acetone 25 ml 58.08 d = 0 .79 g/cm³ PLGA 75/25 0.05 g

CoFe38H 0.02 g

BSA Fraction V 1 g

Synthesis:

5% (w/v) suspension of CoFe38H in acetone are added to the PLGA solution.

(containing PLGA and CoFe38H) to a water flow containing BSA (volume ratio of acetone/water = 1/40). The corresponding immersion tubes draw the solution directly from the tanks containing the two solutions.

Size characterization by means of DLS:

The contrast media comprising the nanometer-sized cobalt ferrite particles which are also an aspect of the present invention, are commonly used contrast media in MRI techniques and normally consist of aqueous possibly isotonic solutions (typically at 0.9% sodium chloride) containing cobalt ferrite up to 20% w/w typically at 1.1 % w/w corresponding to 1 1 mg/ml.

The main experimental result obtained on the various samples with the different measurement types are given hereinafter and can be briefly summarized as follows: relaxometry data: measure the relaxivities R1 and R2 (or the equivalent 1/T₁C and 1/T₂c in solids) at different frequencies (the value of most interest is at around 60 MHz corresponding to H=1.5 Tesla, which is typical of clinical Imagers) on samples at points (a), (b), (c) and (d), at ambient and physiological temperatures. MRI experiments on phantoms containing different sample concentrations at point (d) (and Endorem as comparison).

MRI experiments with a Bruker Avance 200 MHz Imager (4.7 Tesla) on rats, into which Endorem and the samples were injected, at point (d). In this case, the focus was on images of the brain and liver. The discoveries made can be briefly summed up as follows. Relaxometry data on the samples of cobalt ferrite nanoparticles have shown that cobalt ferrite nanoparticles have an R2 signal greater than the iron oxide-based commercial product Endorem.

Relaxometry data on liquid samples have shown that the NBR1 compound has approximately twice the R2 (i.e. efficiency) of Endorem® over the most widely used clinical application range (from 0.2 to 1.5 Tesla). Therefore, where signal loss is important for image contrast (dark areas) the NBR1 compound is a better CA than the most used commercial compound. Using the Artoscan clinical Imager (0.2 Tesla, about 8 MHz) the nano-bioreactor

NBFM _32C C2 was tested in test tubes at different concentrations and compared with Endorem. In confirmation of the relaxometry data, NBR1_32C C2 is more effective than Endorem in lowering the NMR signal (dark areas).

From MRI measurements on rats with a 4.7 Tesla research Imager, and observing the changes in image contrast in the liver and brain consequent to NBR1_32C C2 injection, Endorem and the NBR1_32C C2 compound were found to have similar efficiencies both for systemic injections and intraparenchymal injections.

RELAXOMETRY

The data on the samples consisting of cobalt ferrite nanoparticles dispersed in a liquid phase consisting of glycols show the R2 signal to be about twice that of the iron oxide-based commercial product.

The data on the liquid samples have also shown that the NBR1_32C C2 compound has about twice the R2 of Endorem® (Fig. 1 ). This signifies that, where signal loss is important for image contrast (for example in the case of liver tumours, healthy cells appear to be dark because they have been reached by the negative CA), the NBR1_32C C2 compound is an excellent CA. The E7VVac compound has also demonstrated good transverse relaxivity, R2, comparable with

Endorem® (see Fig. 1 and Table 1 ).

Table 1

MRI experiments on phantoms

In this case the nano-bioreactor NBR1_32C C2 was tested in test tubes at different concentrations and compared with Endorem. As can be seen from the proton density-weighted (PDW) Gradient-Echo (GE) sequences used which correspond to the figures given below, NBR1_32C C2 is more effective than Endorem in lowering the NMR signal (dark areas). In particular, as can be seen in figures 2b and 2c, for concentrations greater than 0.03125 mg Fe/ml, the NBR1_32C C2 signal is lost whereas that of Endorem is still present. Therefore the efficacy of NBR1_32C C2 in negatively contrasting the image is greater, hence confirming the relaxometry characterization data. MRI experiments on rats

As a final step in studying NBR1_32C C2 as a contrast agent, MRI measurements were undertaken on rats (8, for statistical purposes) in order to view changes in the image contrasts of liver and brain as a result of the NBR1_32C C2 injection. Fig. 3 shows some examples of images relating to the case of systemic injection. The data on injected compound NBR1_32C C2 and data on the commercial compound Endorem were compared.

In conclusion it can be stated that Endorem and NBR1_32C C2 at 4.7 Tesla are similarly efficient but NBR1_32C C2 is more effective in negatively contrasting healthy liver cells.

All the images were acquired with gradient echo sequence, te 6 ms, tr 200 ms, flip 30°, 4 averages, 1 mm slices, fov 256x256 points and with in-plane resolution of 200x200 micron. The images acquired of the liver required synchronization with animal respiratory activity. In all the liver images, a reference sample was used and placed next to the rat. Procedure

The CA was injected (systemically) at a concentration of 6 mg Fe/kg. At time zero, the liver appeared dark and the brain showed few traces of CA (Fig. 3a). After 1 day (Fig. 3b), the brain appeared to be emptied of CA while the liver still appeared dark, hence establishing the effectiveness of NBR1_32C C2 for contrasting images of the liver. After 6 days, (Fig. 3c) the liver showed a more marked presence of NBR1 than Endorem (Fig. 3d), i.e. has a greater effectiveness over long periods. The experimental results using NBR1_32C C2 at a 0.075 mg Fe/kg concentration were strictly similar to those of Endorem, thus also confirming a similar tissue response for both compounds as a function of time.

Claims

1. Use of magnetic cobalt ferrite nanoparticles as contrast media in MRI techniques.

2. Use according to claim 1 wherein said nanoparticles are functionalized and/or incorporated in constructs consisting of composite materials.

3. Use according to claims 1 and 2 wherein said nanoparticles have a mean diameter between 3 and 100 nm.

4. Use according to claim 3 wherein said particles are functionalized with mono - and difunctional compounds being compounds of general formula: R₁-(CH₂)_H -R₂ in which: n is an integer between 2 and 20;

R₁ is chosen from: CONHOH, CONHOR, PO(OH)₂, PO(OH)(OR), COOH, COOR,

SH, SR;

5. Use according to claim 1 wherein said nanometer-sized cobalt ferrite particles are incorporated in a water soluble or water insoluble polymer.

6. Use according to claim 5 wherein said water soluble polymers are chosen from: surfactants, polyelectrolytes, water soluble polypeptides and proteins; preferred are water soluble polymers chosen from block copolymers, modified polyethylene glycols, modified polysaccharides, phospholipids, polyaminoamides, globular proteins.

7. Use according to claim 5 wherein said water insoluble polymers are chosen from polyesters, polyamides, polyanhydrides, polyorthoesters, peptides, polyaminoamides, the polymers chosen from polyesters and polyaminoamides being preferred.

8. Use according to claim 7 wherein said insoluble polymers are stabilized with surface active agents chosen from: surfactants, polyelectrolytes, polypeptides and water soluble proteins; preferred are surface active agents chosen from block copolymers, modified polyethylene glycols, modified polysaccharides, phospholipids, polyaminoamides, globular proteins.

9. Contrast agents for MRI techniques comprising nanometer-sized cobalt ferrite particles.

10. Contrast agents according to claim 9 consisting of nanometer-sized cobalt ferrite particles, PLGA and albumin.

1 1. Contrast agents according to claims 9 and 10 as theranostic agents, i.e. comprising both a diagnostic function in MRI and a therapeutic function in magnetic hyperthermia.

12. Contrast agents according to claims 9 and 10 usable for localized and controlled drug delivery by magnetic hyperthermia.