Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print March 20, 2006, 10.1148/radiol.2392042110
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2392042110v1
239/2/415    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fortin-Ripoche, J.-P.
Right arrow Articles by Clément, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fortin-Ripoche, J.-P.
Right arrow Articles by Clément, O.
(Radiology 2006;239:415-424.)
© RSNA, 2006

Experimental Studies

Magnetic Targeting of Magnetoliposomes to Solid Tumors with MR Imaging Monitoring in Mice: Feasibility1

Jean-Paul Fortin-Ripoche, MS, Marie Sophie Martina, MS, Florence Gazeau, PhD, Christine Ménager, PhD, Claire Wilhelm, PhD, Jean-Claude Bacri, PhD, Sylviane Lesieur, PhD and Olivier Clément, MD, PhD

1 From the Laboratoire Matière et Systèmes Complexes, Groupe Physique du Vivant, Université Paris 7, MSC, 140 rue de Lourmel, 75015 Paris, France (J.P.F., F.G., C.W., J.C.B.); Laboratoire de Recherche en Imagerie, Faculté de Médecine Necker, Paris, France (J.P.F., O.C.); Equipe Physico-Chimie des Systèmes Polyphasés, Faculté de Pharmacie, Chatenay-Malabry, France (M.S.M., S.L.); and Laboratoire des Liquides Ioniques et Interfaces Chargées, Université Paris 6, Paris, France (C.M.). Received December 13, 2004; revision requested February 7, 2005; revision received May 2; accepted June 3; final version accepted July 20. Supported by the French Ministry of Education and Research, Centre National de la Recherche Scientifique (ACI Nanoscience et Nanotechnologie NR145), and Institut National de la Santé et de la Recherche Médicale (Programme Interdisciplinaire Imagerie du Petit Animal). Address correspondence to J.P.F. (e-mail: fortin{at}ccr.jussieu.fr).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Purpose: To establish the feasibility of magnetoliposome tumor targeting with an extracorporeal magnet.

Materials and Methods: Animal experiments were performed in compliance with Institut National de la Santé Et de la Recherche Médicale animal protection guidelines and were approved by local government authorities. Magnetophoresis was used to measure the velocity of magnetoliposomes constituted of polyethylene glycol–lipids and anionic maghemite nanocrystals in a calibrated magnetic field in vitro. For in vivo studies, 38 male Swiss nude mice bearing a PC3 human prostate carcinoma tumor in each flank received an intravenous injection of magnetoliposomes (n = 27), saline (n = 9), or nonencapsulated superparamagnetic particles (n = 2) after a small magnet with a magnetic field of 0.3 T and a field gradient of 11 T/m was fixed to the skin above one tumor. The animals were examined at magnetic resonance (MR) imaging with eight different sequences, iron doses (13 mice), and magnet-application durations (12 mice). Their excised tumors were then stained with Perls Prussian blue and hematoxylin-eosin and were examined histologically. With use of the paired Student t test, signal intensity, tumor surface enhancement, and number of stained cells were compared between the control and magnet-exposed tumors to determine significant differences (P ≤ .01).

Results: The mean magnetoliposome velocity ranged from 10 to 40 µm/sec when the magnetic field equaled 0.13 T and the field gradient equaled 25 T/m. At T1-weighted three-dimensional spoiled gradient-echo MR imaging in vivo, the tumor exposed to the magnet showed strong negative enhancement, –52%, compared with the –7% enhancement of the other tumor. Maximal enhancement occurred after 3 hours of magnet application. After 24 hours of magnet application, intracapillary iron particle accumulation was observed in the targeted tumors only.

Conclusion: Magnetic targeting of sterically stabilized magnetoliposomes after they are intravenously injected is feasible in vivo.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
An ongoing challenge in therapeutic procedures is that of limiting the systemic toxic effects by delivering drugs to a precise targeted lesion or region. Biochemical strategies of this type are based on specific interactions between the drug carrier and the targeted lesion or region (1–3), whereas biophysical approaches rely on physical forces to "drag" a responsive vehicle to the target site (4). The particle size, material properties, and route of administration in a targeted drug delivery system determine a given drug's clearance from the reticuloendothelial system, access to the target vasculature, and extravascular uptake (5–7).

The use of liposomes has several advantages in this setting—notably, the potential to coat their surface with repellent molecules (to avoid reticuloendothelial system clearance) or specific ligands (for biologic targeting) and protection of the encapsulated drug substance (8–13). Several strategies to control the release of encapsulated drugs by means of biologic or physical triggers (eg, cell membrane fusion, pH sensitivity, and heat sensitivity) are being developed (14–18).

Another promising approach is to incorporate a colloidal suspension of magnetic nanoparticles—specifically, ferrofluid—into liposomes (19,20). By creating magnetic field inhomogeneity, ferrofluid permits local contrast enhancement in magnetic resonance (MR) imaging (21–23) and thereby offers the potential to monitor magnetoliposomes (MLs) in vivo (18,24). The response of ferrofluid-containing MLs depends on the nature of the magnetic field (25). A high-frequency alternating magnetic field results in local hyperthermia owing to heat generation by the magnetic monodomain nanocrystals. Tumor regression (26) and enhanced antitumor immunity (27) have been observed following magnetically induced hyperthermia. This approach has also been applied to control drug release (28) and transgene expression (29). A nonuniform magnetic field generates a force that is directed toward the larger field, and this force can be used to drive a magnetic carrier toward a target (30). Magnetic drug targeting has been investigated in vivo by using various combinations of magnetic carriers and drugs (4,28,31–36). The purpose of our study was to establish the feasibility of ML tumor targeting with an extracorporeal magnet.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Magnetic Nanoparticles
The magnetic fluid that we used was composed of monodisperse nanocrystals of maghemite that was synthesized according to the method of Massart (37) and Lefébure et al (38). The stability of the particles was ensured by the specific adsorption of citrate ions to their surface, which caused the particles to have a negative charge. The nanocrystals—dispersed in a buffer composed of 10 mmol/L of 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid, 108 mmol/L of sodium chloride, and 20 mmol/L of sodium citrate (pH 7.4, 285 mOsm)—consisted of ferrimagnetic monodomains with a mean magnetic diameter of 7.5 nm ± 0.3 (standard deviation).

ML Preparation and Characterization
The sterically stabilized magnetic fluid–loaded liposomes were composed of 95 mol of egg phosphatidylcholine plus 5 mol of 1.2-diacyl-SN-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (DSPE-PEG2000; Avanti, Alabaster, Ala) and were prepared by using lipid film hydration followed by sequential extrusion, as previously described (14,39). Nonencapsulated magnetic nanocrystals were removed by using gel exclusion chromatography. Cryotransmission electron microscopy revealed mostly unilamellar spherical vesicles containing nonaggregated maghemite nanocrystals. The mean hydrodynamic diameters of the MLs, which ranged from 150 to 250 nm, were determined by using quasielastic light scattering with a nanosizer (Coulter Electronics, Harpenden, United Kingdom). The total lipid concentration was 20 mmol/L, and 1.65 mol of iron per mole of lipid was entrapped, according to flame spectroscopy measurements. The preparations were stable for at least 6 months. The magnetization curve of the ML dispersions showed superparamagnetic behavior without hysteresis and was characterized by 50% of the magnetization saturating value at a magnetic field strength of 0.04 T and by 80% of the magnetization saturating value at 0.17 T. The ML r1 and r2 relaxivities at 20 MHz (0.47 T) and 37°C measured by using a spectrometer (Minispec PC120; Bruker, Rheinstetten, Germany) were 7.65 L · mmol–1 · sec–1 and 130 L · mmol–1 · sec–1, respectively, according to variations in T1 and T2 relaxation times as a function of the iron concentration.

Magnetophoretic Mobility of MLs
In a nonuniform magnetic field (B), an ML bearing a magnetic moment (m) is subjected to a torque (m · B), which aligns the ML with the field, and to a magnetophoretic driving force (FM = G[m · B], where G is the gradient operator), which moves the ML toward the stronger field.

Two magnetophoretic experiments were conducted to characterize the magnetophoretic mobility of a ML suspended in a given field geometry. In the magnetophoretic setup (40), ML movement was observed along the axis of a circular permanent magnet that was developing a field of 0.17 T and a gradient of 18.5 T/m colinear to each other. In the micromagnetophoretic setup, MLs were placed between two rectangular permanent magnets that had developed a uniform parallel magnetic field of 0.1 T. Two 50-µm-diameter nickel rods that were aligned with the field lines and separated by a 400-µm gap were used to produce a local deformation of the magnetic field. On a (100 µm)2 zone centered in the middle of the setup, this deformation corresponded to a gradient of 6.8 T/m perpendicular to the magnetic field. In the magnetophoretic and micromagnetophoretic setups, the MLs had a constant magnetophoretic velocity when the hydrodynamic drag force balanced the magnetic force. ML movement was recorded at videomicroscopy, and the separate velocities of 200 MLs (magnetophoresis setup) and of 100 MLs (micromagnetophoresis setup) were measured by using image processing.

For the in vivo tests, the magnet used consisted of a circular cylinder (radius, 7 mm; height, 5 mm; weight, 5 g) made of neodymium, iron, and boron. The magnetic field at the magnet surface was 0.29 T. The magnetic field amplitude and gradient were measured along the axis of revolution by using a Hall effect probe. Then, the ML velocities within 5 mm of the magnet were extrapolated from these measurements.

Animal Model
The animal experiments were compliant with Institut National de la Santé Et de la Recherche Médicale animal protection guidelines and approved by local government authorities. The experiments were performed by using 38 male Swiss nude mice (Iffa Credo, l'Arbresle, France) that weighed between 25 and 31 g; these animals are amenable to tumor implantation because of their immunosuppression. We anesthetized the mice by injecting them intraperitoneally with a xylazine 2% (Rompum; Bayer, Leverkusen, Germany) and ketamine (Imalgene 500; Rhône Mérieux, Lyon, France) solution (0.1 mL per 10 g of body weight, 4:1 vol/vol). The normal body temperature of the anesthetized animals was maintained by using a heating lamp.

Human prostatic adenocarcinoma cells (provided by M.F. Poupon, Institut Curie, Paris, France) were cultured in Dulbecco modified Eagle medium supplemented with L-glutamine (Glutamax), 10% fetal calf serum, and 100 IU/mL of penicillin-streptomycin. (All cell culture products were from Gibco/Invitrogen, Cergy Pontoise, France.) When confluence was achieved, the cells were treated with trypsin and counted. Suspensions of 1.5 x 106 cells in 0.2 mL of phosphate-buffered saline were injected subcutaneously into the right and left flanks, at the level of the heart. The choice of this site minimized animal discomfort and reduced breathing artifacts during MR imaging. Two tumors, measuring 7 x 7 mm on average, developed after 25 days.

Magnetic Targeting in Vivo
The mice were anesthetized as described earlier. Cyanoacrylate glue was then used to stick a magnet (Fig 1, top) to the skin above one tumor, and a containing band (Elastoplaste; BSN Medical, Le Mans, France) was bound around both tumors. The ML suspension (100 or 200 µL) was then injected into one caudal vein, and the magnet was left in place for 1–24 hours. The tumor juxtaposing the magnet was referred to as the targeted tumor, and the other tumor was referred to as the control tumor.


Figure 1
View larger version (38K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Top: Schematic illustration of the external magnet applied over the tumor. The top (black) line represents the measured magnetic field along the axis, and the bottom (gray) line represents the calculated gradient. The magnetic field is higher than 0.05 T to a depth of 10 mm, indicating substantial magnetization of the MLs. The magnetic gradient exceeds 10 T/m over the same distance, which is about the size of the tumor. Bottom: Graph illustrates ML velocities 5 mm from the magnet in vivo. This distribution was extrapolated from the magnetophoresis experiments.

 
Thirty-eight animals were divided into three groups according to the injected material. The targeted delivery group comprised 27 animals injected with MLs. The reference group comprised nine mice injected with saline, and the superparamagnetic particle group comprised two mice injected with nonencapsulated superparamagnetic particles (Sinerem; Guerbet, Aulnay sous Bois, France).

The targeted delivery group was further divided into three subgroups: The first subgroup was used for an exposure time study and comprised 12 mice that bore the magnet for 1 (n = 3), 2 (n = 1), 3 (n = 1), 4 (n = 5), or 5 (n = 2) hours before being examined with MR imaging. The second subgroup was used for an iron dose study and comprised 13 mice that received ML doses of 4.3–14.7 mg of iron per kilogram of body weight. The third group was used for a drug release study and comprised two mice in which the magnet was left in place for 1 hour following ML injection (3.0 mg/kg of iron) and that were examined with MR imaging immediately and 16 hours later. The experimental parameters and animal groups are summarized in Table 1.


View this table:
[in this window]
[in a new window]

  		Table 1. Definition and Experiment Characteristics of Animal Groups

 
MR Imaging and Evaluation
MR imaging was performed with a 1.5-T clinical MR device (Signa 1.5 T; GE Medical Systems, Milwaukee, Wis). A custom-built 3-cm-diameter Helmholtz coil was used for emission and reception.

After being anesthetized, the animals were placed in the coil, which contained an oil tube that served as a reference phantom. T1-weighted spin-echo (500/11 [repetition time msec/echo time msec]), T2-weighted spin-echo (2000/20, 40, 60, 80), T2-weighted gradient-echo (140/10–25, 30° flip angle), and three-dimensional (3D) spoiled gradient-echo (SPGR) (37.4/8.6, 30° flip angle) sequences were performed with a 4 x 4-cm field of view, a 256 x 128 matrix, one acquired signal, and a 1-mm (SPGR sequence) or 2-mm (all other sequences) section thickness. The MR imaging parameters are summarized in Table 2.


View this table:
[in this window]
[in a new window]

  		Table 2. MR Imaging Sequence Characteristics

 
The effects of ML accumulation on MR imaging signal intensity were quantified by measuring the intensity of the enhancement and the percentage of enhanced cross-sectional tumor area. Signal intensity was first normalized to the oil phantom signal intensity on the same image and is referred to as the normalized signal intensity. Then, the enhancement (E) between the normalized signal intensity in the ML-injected animals (SInor) and the mean of the reference signal intensity values (SIref) was calculated as a percentage for each region of interest (ROI) as follows: E = [(SInor – SIref)/SIref] · 100. For all animals and sequences, a single observer (J.P.F.) blinded to the mouse groups defined the following six oval ROIs with the aid of imaging software (Advantage Windows; GE Medical Systems): control tumor, targeted tumor, blood (in left ventricle), liver, spleen, and muscle. Because of the heterogeneous enhancement of the targeted tumors, an additional oval ROI covering one-fifth of the tumor area was placed on the maximally enhanced zone in the targeted tumor (TTmax). The control tumors were more homogeneous and did not require this seventh ROI study. Enhancement values for the control tumor, targeted tumor, and TTmax were obtained by drawing one ROI on three or four representative image sections. Then, the surface area (SROI) and enhancement (EROI) of these regions were measured and the overall enhancement (E) was calculated by weighting the enhancement value for each ROI by the surface area of the region, as follows:

Formula

Because of the heterogeneity of negative enhancement of the targeted tumors, we tried to characterize the spatial extension of the MLs. For this purpose, the proportion of the tumor surface area in which the signal intensity was lower than 80% of the nonenhanced tumor signal intensity was determined. Each 3D SPGR transverse image section that included the tumor was normalized to the signal intensity of the nonenhanced part of the tumor. The surface areas of the tumor (St) and of the part of the tumor in which the normalized pixel value was lower than 80% (Senh) were measured by using computer software (Scion Image; Scion, National Institutes of Health, Bethesda, Md). Then, the proportion of the tumor surface area negatively enhanced by more than –20% (PSenh) was computed: PSenh = Senh/St. For each tumor, this analysis was performed on three or four representative sections, and the mean PSenh was calculated.

The influence of the magnet application duration on ML uptake was assessed by plotting for each ROI the mean enhancement at 3D SPGR MR imaging for the five exposure times (Fig 2). The mice used in the drug release study were analyzed for their PSenh after 1 hour of magnet exposure and then 16 hours later.


Figure 2
View larger version (30K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2: Graph illustrates mean enhancement values for different organs on 3D SPGR MR images as a function of magnet exposure duration. The targeted tumor (TT) uptake of MLs gradually reached a maximum after 3 hours of magnet application. Blood showed negative enhancement that persisted for 5 hours, indicating that MLs were still circulating in vessels. Muscle showed no marked ML uptake. The liver and spleen were negatively enhanced by more than –80% during the first hour because of the nonspecific uptake of MLs. * = significant difference in maximal enhancement between targeted tumors and control tumors (CT) after 4 hours of magnet application in the five mice used in this experiment (P ≤ .01, paired Student t test). For improved clarity, error bars are not shown.

 
Histologic Analyses
The left and right tumors, the spleen, and the liver were surgically harvested and fixed in acetic formalin solution. Six-micrometer slices were stained with hematoxylin-eosin and Perls Prussian blue and then examined with bright-light microscopy.

The stained cells and capillaries were counted on 65 slices from 16 tumors (eight targeted, eight control) in the targeted delivery group and on 16 slices from four tumors in the superparamagnetic particle group by a single observer (J.P.F.) who was blinded to the injected material and as to whether or not the tumor had been subjected to magnetic exposure. To characterize the stained cells and determine whether the glue or the magnet induced an inflammatory reaction, MAC3 macrophage immunolabeling was performed in two mice.

Statistical Analyses
The mean enhancement values (± standard deviations) for the targeted tumor, control tumor, and TTmax in the targeted delivery and superparamagnetic particle groups were computed for each MR imaging sequence. The normality of the targeted tumor and control tumor enhancement values was determined with the Shapiro-Wilks test. Then, targeted tumor enhancement values were compared with control tumor enhancement values in the same animal by using the paired Student t test (P ≤ .01).

The same statistical tests were used to compute targeted and control tumor ROI enhancement values in the 4-hour (magnet application) group (relating only to the 3D SPGR data). In the iron dose study group, the Pearson correlation between the injected dose and the negative enhancement of the targeted tumor was computed.

The mean PSenh (± standard deviation) for the control and targeted tumors in the targeted delivery group was computed. Normality was checked with the Shapiro-Wilks test, and the targeted tumor PSenh was compared with the control tumor PSenh in the same animal by using the Student paired t test (P ≤ .01).

The mean number of labeled cells (± standard deviation) was computed for the targeted and control tumors in the targeted delivery and superparamagnetic groups. Normality was checked with the Shapiro-Wilks test, and the number of labeled cells in the targeted tumor was compared with that in the control tumor in each animal by using the paired Student t test (P ≤ .01). The Pearson correlation between the number of stained capillaries on histologic sections of targeted tumor and the number of stained capillaries on histologic sections of TTmax was computed. These analyses were performed by using 2002 Excel software (Microsoft, Redmond, Wash).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Magnetophoretic Mobility
In the micromagnetophoretic and magnetophoretic setups, mean velocities (Fig 3) were 2.87 µm/sec ± 0.93 and 17.7 µm/sec ± 6.08, respectively. The two distributions had the same shape, with a standard deviation equal to one-third of the mean and the mean velocity differing by a factor of six. The shift in magnetophoretic mobility was mainly due to the fact that the magnetic field and the magnetic field gradient were different in the two setups. In addition, mobile MLs, which were elongated along the field direction, experienced a larger hydrodynamic friction force when they moved perpendicular to the field (micromagnetophoretic setup) than when they moved parallel to the field (magnetophoretic setup).


Figure 3
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3: Graph illustrates distribution of ML velocity in two experimental conditions. In the micromagnetophoresis experiment (dark gray bars), 100 MLs were submitted to a magnetic field of 0.1 T and a gradient of 6.8 T/m perpendicular to each other. In the magnetophoresis experiment involving 200 MLs (light gray bars), the magnetic field (0.17 T) and the gradient (18.5 T/m) were colinear. The mean velocity was four times lower in the micromagnetophoresis experiment because of the threefold smaller field gradient and orientation of the MLs perpendicular to their displacement. The scale is discontinuous. Values on the vertical axis are percentages of the total number of liposomes observed.

 
The magnetic field and field gradient of the magnet used in vivo are illustrated in Figure 1 (top) as a function of the distance from the magnet surface. The field intensity was maximal (0.29 T) at the magnet surface and remained higher than 0.05 T more than 10 mm from the magnet surface. The field gradient was maximal (30 T/m) 1 mm from the magnet surface and almost constant up to 3 mm from the surface.

At 5 mm from the magnet surface, the magnetic field was 0.13 T and the field gradient was 25 T/m. The distribution of ML velocities (Fig 1, bottom) was extrapolated from the magnetophoretic mobility experiments. ML velocities ranged from 10 to 40 µm/sec 5 mm from the magnet in vivo and from 2 to 70 µm/sec in the tumor itself.

MR Imaging Monitoring of ML Tumor Targeting in Vivo
A clear heterogeneous darkening could be qualitatively observed on the tumors exposed to the magnet (Fig 4). The difference between targeted and control tumors was more pronounced on the gradient-echo images (Fig 4, C, D) than on the spin-echo images (Fig 4, A, B, B', B"), with the T2-weighted spin-echo images being more sensitive to MLs.


Figure 4
View larger version (139K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4: Transverse post–ML injection T1-weighted spin-echo (500/11) (A), density-weighted spin-echo (2000/20, 40, 60, 80) (B, B'), strongly T2-weighted spin-echo (2000/20, 40, 60, 80) (B"), gradient-echo (140/10–25) (C), and 3D SPGR (37.4/8.6) (D) MR images obtained in a mouse with a subcutaneously implanted tumor in each flank. An oil phantom was placed on the animal's back. The magnet was glued to the skin over the left tumor for 24 hours after intravenous injection of the MLs (260 µmol/kg of iron). The magnet was removed before imaging. The magnetically targeted tumor (left) appears darker than the contralateral (control) tumor (right), indicating preferential accumulation of MLs. The darkening is more pronounced on B", C, and D than on A, B, and B'.

 
Quantitative analysis of the enhancement of targeted tumors, TTmax regions, and control tumors in the targeted delivery group (Fig 5) revealed that both the targeted tumors and the TTmax regions were markedly darker than the control tumors. All targeted and control tumor enhancement values at the different sequences were normally distributed. The differences in enhancement between the control and targeted tumors were significant at all sequences (P ≤ .01) except the T2-weighted sequence performed with 60- and 80-msec echo times. The darkening effect was clearly sequence dependent, increasing with increasing echo time in the spin-echo sequences and being more pronounced in the gradient-echo and 3D SPGR sequences. The 3D SPGR MR images were used for subsequent analyses because of their better definition and contrast properties and more marked depiction of the difference between targeted and control tumors.


Figure 5
View larger version (29K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5: Graph illustrates mean tumor enhancement values (± standard deviations) as a function of the MR imaging sequence (A = T1-weighted spin-echo [500/11], B = T2-weighted spin-echo [2000/20, 40, 60, 80], C = T2-weighted gradient-echo [140/10–25], D = 3D SPGR [37.4/8.6]; numbers in parentheses are echo times). * = significant difference (P ≤ .01) between targeted (TT) and control (CT) tumors. Only those animals with two visible tumors were analyzed.

 
In the animals injected with superparamagnetic particles, enhancement of the targeted tumors was slightly more pronounced than enhancement of the control tumors (Table 3). However, the qualitative aspect of the MR images and the order of magnitude of the magnetic force led us to conclude that magnetic targeting on 7-nm-diameter magnetic particles was inefficient.


View this table:
[in this window]
[in a new window]

  		Table 3. Tumor Enhancement Values in Two Mice in Superparamagnetic Particle Group

 
With regard to enhanced surface area values (Fig 6), the mean PSenh of the targeted tumors in the 27 mice injected with MLs was 47.9% ± 19.1 (standard deviation). The mean PSenh of the control tumors in these mice was 14.8% ± 9.7. This difference was significant (P ≤ .01).


Figure 6
View larger version (25K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6: Graph illustrates PSenh values on 3D SPGR MR images (37.4/8.6, 30° flip angle). The control tumors (CT) showed ML accumulation on less than 30% of the surface area. The magnet-exposed (targeted) tumors (TT) exhibited a larger enhanced surface area: 15%–85%.

 
With regard to mean enhancement as a function of magnet exposure duration after ML injection (Fig 2), three curve shapes were observed: (a) a marked decrease in enhancement during the first hour followed by an enhancement plateau (negative value) in the liver, spleen, and blood; (b) small variations in muscle and control tumor enhancement; and (c) a gradual, almost linear negative enhancement of the targeted tumor and the TTmax. Maximal negative enhancement of the targeted tumor (–33.7%) and the TTmax (–63.0%) was reached after 3 hours of magnet exposure. The injected dose correlated linearly with targeted tumor enhancement (r = –0.85).

In the drug release group, a slight difference in the PSenh of the targeted and control tumors was detected between the images obtained after 1 hour of magnet exposure (targeted tumor, 65.5%; control tumor, 13.9%) and those obtained 16 hours after magnet removal (targeted tumor, 50.9%; control tumor, 13.8%).

Histologic Analyses
Histologic analysis of the tumors revealed a central necrotic region and a vascularized rim. On tumor cross sections, a blue coloration was detected either in isolated cells in the tumor interstitium (Fig 7a) or in ML clusters in capillaries (Fig 7b). Capillaries with numerous intraluminal blue inclusions were found in targeted tumors but rarely in control tumors. The graph in Figure 8 illustrates the correlation between number of stained capillaries and signal intensity enhancement at 3D SPGR MR imaging for the control and targeted tumors (Pearson correlation coefficient, –0.75).


Figure 7
View larger version (175K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7a: Histologic analysis of two stained slices of targeted tumor. Blue coloration indicates the presence of iron. (a) Marked accumulation of MLs is seen at the tumor periphery, in the most highly vascularized zone. (Perls Prussian blue and hematoxylin-eosin stain; magnification, x10.) (b) Uniform accretion of ML is seen at the inner walls of the capillaries, suggesting ML attraction by means of magnetic cooperation. (Perls Prussian blue and hematoxylin-eosin stain; magnification, x100.)

 

Figure 7
View larger version (136K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 7b: Histologic analysis of two stained slices of targeted tumor. Blue coloration indicates the presence of iron. (a) Marked accumulation of MLs is seen at the tumor periphery, in the most highly vascularized zone. (Perls Prussian blue and hematoxylin-eosin stain; magnification, x10.) (b) Uniform accretion of ML is seen at the inner walls of the capillaries, suggesting ML attraction by means of magnetic cooperation. (Perls Prussian blue and hematoxylin-eosin stain; magnification, x100.)

 

Figure 8
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 8: Graph illustrates correlation between number of labeled vascular capillaries and absolute enhancement. The curve shows the relationship between the vascular pattern of iron accretion observed in the targeted tumors (see Fig 7b) and the negative enhancement associated with ML uptake.

 
In the targeted delivery group, the mean frequency of Perls Prussian blue and hematoxylin-eosin–stained cells per slice was significantly higher in the targeted tumors (53.3 cells per slice ± 39.1) than in the control tumors (5.96 cells per slice ± 5.0) (P ≤ .01). No such difference was found in the superparamagnetic particle group (8.63 cells per slice in targeted tumors, 2.25 cells per slice in control tumors). In the targeted tumors, the labeled cells consisted equally of fibroblasts, macrophages, and endothelial cells; no tumor cell labeling was observed. Iron deposits were found in numerous cells in the liver and spleen.

About 50% of the macrophages were also stained, indicating that not all MLs had been taken up by macrophages 24 hours after the MLs were injected. The relative number of macrophages was normal, demonstrating the absence of an inflammatory reaction.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
The concept of magnetic targeting (31,41) was introduced by Widder et al (31) in 1978. Several pharmacologically stable formulations associating a drug with a magnetically active component have since been developed, but few have proved to be effective in animal models. A magnetic targeted carrier bound to doxorubicin (MTC-DOX) (metallic iron-activated carbon–doxorubicin; Ferx, San Diego, Calif) is a composite 0.5–5.0-µm-diameter microparticle that is composed of metallic iron and activated carbon and loaded with doxorubicin. Intraarterially infused MTC-DOX has been efficiently targeted to hepatocellular carcinomas with the aid of an external magnet both in animal models (42) and in phase I and II clinical trials (35,43). Using an intravenously injected ferrofluid (particle size, 100 nm) chemically bound to epirubicin, Lübbe et al (32,44) and Lemke et al (45) reported that after 60–120 minutes of magnetic field application to the tumor site, the ferrofluid was successfully directed to the tumor, with good tolerance, in half the patients examined. Colloidal liposomes 200 nm in diameter are more recently discovered carriers that have specific advantages for drug targeting and delivery.

Liposomes are vesicular systems in which a drug is confined to an aqueous cavity surrounded by a membrane, which protects the drug from premature inactivation. In the current study, we used magnetic nanoparticles encapsulated in the liposome cavity. Intravenous administration of ferrofluid (superparamagnetic particles) did not lead to preferential accumulation in the magnet-exposed tumor. Confinement of magnetic nanoparticles to the liposome cavity yields a global magnetic moment that is sufficient to drive the liposomes against the blood flow with use of an external magnet. The ML velocity distribution is of the same order of magnitude as the blood velocity distribution in capillary vessels (46).

Another key advantage of liposomal encapsulation is that the biodistribution of MLs is determined according to their size and membrane properties instead of the inherent properties of the magnetic nanocrystals. The phospholipid bilayer that constitutes the ML membrane is coated with a hydrophilic polymer, polyethylene glycol, creating a dynamic "cloud" of hydrophilic neutral chains at the ML surface, which are expected to repel plasma proteins and thereby reduce opsonization reactions and the subsequent clearance by macrophages. Our study results show that a fraction of the polyethylene glycol–stabilized MLs circulated for a long enough time to selectively extravasate into tumors, which were characterized by a leaky vasculature. Thus, use of polyethylene glycol–stabilized liposomes enables one to overcome the limitations of intraarterial injections encountered in most efficient in vivo magnetic targeting studies. Their submicrometric size (200 nm) facilitates tumor targeting by favoring access to the tumor interstitium and interactions with tumor cells. Hence, the mechanism of ML accumulation in tumor tissue relies first on the passive diffusion through the hyperpermeable vasculature.

We compared the tumor accumulation of MLs between two tumors in the same animals, with one of the tumors being exposed to an external magnet. The MR contrast properties of MLs enable the particles to be monitored noninvasively in vivo after they are intravenously injected. Massive uptake by the reticuloendothelial system (especially the spleen and liver) was clearly indicated by the darkening of these organs after 1 hour. Blood pool imaging revealed a long plasma half-life. Muscle did not show marked ML uptake. The magnet-exposed tumors showed significantly higher enhancement than the contralateral control tumors, and this enhancement was proportional to the ML dose. The distribution of MLs in the magnet-exposed tumors was preferential in the vicinity of the magnet surface and in the vascular region. The persistence of marked enhancement 16 hours after magnet removal indicates that MLs were not released after tumor capture, and this finding represents another advantage over the larger magnetic particles (100 nm to 1 µm) used in other studies of magnetic targeting.

Perls Prussian blue–hematoxylin-eosin staining of histologic sections revealed passive ML deposition in a few macrophages of the control tumors. In contrast, the magnet-exposed tumors contained ML aggregates on the capillary walls. Finally, the nondetection of an inflammatory reaction in the targeted tumors shows that the ML accumulation was not due to tumor invasion by ML-loaded macrophages and confirms the physical origin of the ML accretion.

There were limitations to our study. The uptake of MLs in the liver is not an issue with regard to imaging, but it is an important factor in therapy. Moreover, our histologic analysis 24 hours after ML injection revealed no intimate interactions between the MLs and the tumor cells. The use of an extracorporeal magnet limits the application of this technique to superficial tumors. Improvements in magnetic field intensity and geometry will be required to treat deeply imbedded tumors (47).

In conclusion, we characterized the magnetophoretic mobility of 200-nm-diameter MLs in vitro and in vivo. Efficient magnetic targeting of MLs toward tumor capillaries, followed by mild retention in the tumor interstitium, was achieved by using an external magnet. The distribution of the MLs was monitored noninvasively with MR imaging. Further studies of anticancer drugs contained in the liposome cavity and of magnetic hyperthermia-controlled drug release are now required (17,26).

Practical applications: After improvements in the biodistribution and specificity of submicrometric polyethylene glycol–stabilized MLs are made, a toxicologic study to obtain clinical approval should be conducted. Then, magnetic targeting could be a relevant alternative to using the larger carrier systems currently being assessed in clinical trials. Use of the described magnetic targeting approach would offer a longer residence time for drugs after they are intravenously injected, limit the risk of embolism, and be suitable for new strategies of magnetically controlled drug release. It is important to note that ML-based drug delivery can be monitored with MR imaging (45).


   ACKNOWLEDGMENTS
 
The authors thank Charles-André Cuenod, MD, PhD, Nathalie Siauve, MD, PhD, and Pierre Smirnov, MS for helpful discussions and Jacques Bittoun, MD, PhD, Patrick Bruneval, MD, PhD, Catherine Vayssettes, PhD, and Josette Prudhomme for their assistance.


   FOOTNOTES
 

Abbreviations: ML = magnetoliposome • PSenh = proportion of the tumor surface area negatively enhanced by more than –20% • ROI = region of interest • SPGR = spoiled gradient echo • TTmax = maximally enhanced zone in the targeted tumor • 3D = three-dimensional

Author contributions: Guarantors of integrity of entire study, J.P.F., F.G., C.W., O.C.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, J.P.F., M.S.M., F.G., C.M.; experimental studies, J.P.F., M.S.M., F.G., C.M., C.W., S.L.; statistical analysis, J.P.F., O.C.; and manuscript editing, J.P.F., M.S.M., F.G., C.M., C.W., S.L., O.C.

Authors stated no financial relationship to disclose.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Marcucci F, Lefoulon F. Active targeting with particulate drug carriers in tumor therapy: fundamentals and recent progress. Drug Discov Today 2004;9:219–228.[CrossRef][Medline]
  2. Crommelin DJ, Scherphof G, Storm G. Active targeting with particulate carrier systems in the blood compartment. Adv Drug Deliv Rev 1995;17:49–60.
  3. Maruyama K, Ishida O, Takizawa T, Moribe K. Possibility of active targeting to tumor tissues with liposomes. Adv Drug Deliv Rev 1999;40:89–102.[CrossRef][Medline]
  4. Hafeli UO. Magnetically modulated therapeutic systems. Int J Pharm 2004;277:19–24.[CrossRef][Medline]
  5. Torchilin VP. Drug targeting. Eur J Pharmacol Sci 2000;11(suppl 2):S81–S91.[CrossRef][Medline]
  6. Jang SH, Wientjes MG, Lu D, Au JL. Drug delivery and transport to solid tumors. Pharm Res 2003;20:1337–1350.[CrossRef][Medline]
  7. Yuan F, Leunig M, Huang SK, Berk DA, Papahadjopoulos D, Jain RK. Microvascular permeability and interstitial penetration of sterically stabilized (stealth) liposomes in a human tumor xenograft. Cancer Res 1994;54:3352–3356.[Abstract/Free Full Text]
  8. Woodle MC. Sterically stabilized liposome therapeutics. Adv Drug Deliv Rev 1995;16:249–265.[CrossRef]
  9. Oku N. Anticancer therapy using glucuronate modified long-circulating liposomes. Adv Drug Deliv Rev 1999;40:63–73.[CrossRef][Medline]
  10. Maruyama K. PEG-immunoliposome. Biosci Rep 2002;22:251–266.[CrossRef][Medline]
  11. Hodenius M, De Cuyper M, Desender L, Muller-Schulte D, Steigel A, Lueken H. Biotinylated stealth magnetoliposomes. Chem Phys Lipids 2002;120:75–85.[CrossRef][Medline]
  12. Barratt G. Colloidal drug carriers: achievements and perspectives. Cell Mol Life Sci 2003;60:21–37.[CrossRef][Medline]
  13. Papahadjopoulos D, Allen TM, Gabizon A, et al. Sterically stabilized liposomes: improvements in pharmacokinetics and antitumor therapeutic efficacy. Proc Natl Acad Sci U S A 1991;88:11460–11464.[Abstract/Free Full Text]
  14. Lesieur S, Grabielle-Madelmont C, Menager C, et al. Evidence of surfactant-induced formation of transient pores in lipid bilayers by using magnetic-fluid-loaded liposomes. J Am Chem Soc 2003;125:5266–5267.[CrossRef][Medline]
  15. Viroonchatapan E, Sato H, Ueno M, Adachi I, Tazawa K, Horikoshi I. Release of 5-fluorouracil from thermosensitive magnetoliposomes induced by an electromagnetic field. J Control Release 1997;46:263–271.
  16. Yuyama Y, Tsujimoto M, Fujimoto Y, Oku N. Potential usage of thermosensitive liposomes for site-specific delivery of cytokines. Cancer Lett 2000;155:71–77.[CrossRef][Medline]
  17. Eeckman F, Moes AJ, Amighi K. Surfactant induced drug delivery based on the use of thermosensitive polymers. J Control Release 2003;88:105–116.[CrossRef][Medline]
  18. Viglianti BL, Abraham SA, Michelich CR, et al. In vivo monitoring of tissue pharmacokinetics of liposome/drug using MRI: illustration of targeted delivery. Magn Reson Med 2004;51:1153–1162.[CrossRef][Medline]
  19. Bogdanov J, Alexei A, Martin C, Weissleder R, Brady TJ. Trapping of dextran-coated colloids in liposomes by transient binding to aminophospholipid: preparation of ferrosomes. Biochim Biophys Acta Biomembr 1994;1193:212–218.[Medline]
  20. Bulte JW, de Cuyper M, Despres D, Frank JA. Preparation, relaxometry, and biokinetics of PEGylated magnetoliposomes as MR contrast agent. J Magnetism Magn Mater 1999;194:204–209.[CrossRef]
  21. Okuhata Y. Delivery of diagnostic agents for magnetic resonance imaging. Adv Drug Deliv Rev 1999;37:121–137.[CrossRef][Medline]
  22. Tilcock C. Delivery of contrast agents for magnetic resonance imaging, computed tomography, nuclear medicine and ultrasound. Adv Drug Deliv Rev 1999;37:33–51.[CrossRef][Medline]
  23. Bulte JW, De Cuyper M. Magnetoliposomes as contrast agents. Methods Enzymol 2003;373:175–198.[Medline]
  24. Pauser S, Reszka R, Wagner S, Wolf KJ, Buhr HJ, Berger G. Liposome-encapsulated superparamagnetic iron oxide particles as markers in an MRI-guided search for tumor-specific drug carriers. Anticancer Drug Des 1997;12:125–135.[Medline]
  25. Halbreich A, Roger J, Pons JN, et al. Biomedical applications of maghemite ferrofluid. Biochimie 1998;80:379–390.[Medline]
  26. Ito A, Tanaka K, Kondo K, et al. Tumor regression by combined immunotherapy and hyperthermia using magnetic nanoparticles in an experimental subcutaneous murine melanoma. Cancer Sci 2003;94:308–313.[CrossRef]
  27. Suzuki M, Shinkai M, Honda H, Kobayashi T. Anticancer effect and immune induction by hyperthermia of malignant melanoma using magnetite cationic liposomes. Melanoma Res 2003;13:129–135.[CrossRef][Medline]
  28. Viroonchatapan E, Sato H, Ueno M, Adachi I, Tazawa K, Horikoshi I. Magnetic targeting of thermosensitive magnetoliposomes to mouse livers in an in situ on-line perfusion system. Life Sci 1996;58:2251–2261.[CrossRef][Medline]
  29. Ito A, Shinkai M, Honda H, Kobayashi T. Heat-inducible TNF-alpha gene therapy combined with hyperthermia using magnetic nanoparticles as a novel tumor-targeted therapy. Cancer Gene Ther 2001;8:649–654.[CrossRef][Medline]
  30. Asmatulu R, Zalich MA, Claus RO, Riffle JS. Synthesis, characterization and targeting of biodegradable magnetic nanocomposite particles by external magnetic fields. J Magnetism Magn Mater 2005;292:108–119.[CrossRef]
  31. Widder KJ, Morris RM, Poore G, Howard DP Jr, Senyei AE. Tumor remission in Yoshida sarcoma-bearing rats by selective targeting of magnetic albumin microspheres containing doxorubicin. Proc Natl Acad Sci U S A 1981;78:579–581.[Abstract/Free Full Text]
  32. Lubbe AS, Alexiou C, Bergemann C. Clinical applications of magnetic drug targeting. J Surg Res 2001;95:200–206.[CrossRef][Medline]
  33. Kuznetsov AA, Filippov VI, Alyautdin RN, Torshina NL, Kuznetsov OA. Application of magnetic liposomes for magnetically guided transport of muscle relaxants and anti-cancer photodynamic drugs. J Magnetism Magn Mater 2001;225:95–100.[CrossRef]
  34. Jain S, Mishra V, Singh P, Dubey PK, Saraf DK, Vyas SP. RGD-anchored magnetic liposomes for monocytes/neutrophils-mediated brain targeting. Int J Pharm 2003;261:43–55.[CrossRef][Medline]
  35. Wilson MW, Kerlan RK Jr, Fidelman NA, et al. Hepatocellular carcinoma: regional therapy with a magnetic targeted carrier bound to doxorubicin in a dual MR imaging/conventional angiography suite—initial experience with four patients. Radiology 2004;230:287–293.[Abstract/Free Full Text]
  36. Nobuto H, Sugita T, Kubo T, et al. Evaluation of systemic chemotherapy with magnetic liposomal doxorubicin and a dipole external electromagnet. Int J Cancer 2004;109:627–635.[CrossRef][Medline]
  37. Massart R. Preparation of aqueous magnetic liquids in alkaline and acidic media. IEEE Trans Magn 1981;17:1247–1248.[CrossRef]
  38. Lefébure S, Dubois E, Cabuil V, Neveu S, Massart R. Monodisperse magnetic nanoparticles: preparation and dispersion in water and oils. J Mater Res 1998;13:2975–2981.
  39. Martina MS, Fortin JP, Menager C, et al. Generation of superparamagnetic liposomes revealed as highly efficient MRI contrast agents for in vivo imaging. J Am Chem Soc 2005;127(30):10676–10685.[CrossRef][Medline]
  40. Wilhelm C, Gazeau F, Bacri JC. Magnetophoresis and ferromagnetic resonance of magnetically labeled cells. Eur Biophys J 2002;31:118–125.[CrossRef][Medline]
  41. Driscoll CF, Morris RM, Senyei AE, Widder KJ, Heller GS. Magnetic targeting of microspheres in blood flow. Microvasc Res 1984;27:353–369.[CrossRef][Medline]
  42. Leakakos T, Ji C, Lawson G, Peterson C, Goodwin S. Intravesical administration of doxorubicin to swine bladder using magnetically targeted carriers. Cancer Chemother Pharmacol 2003;51:445–450.[Medline]
  43. Goodwin S, Peterson C, Hoh C, Bittner C. Targeting and retention of magnetic targeted carriers (MTCs) enhancing intra-arterial chemotherapy. J Magnetism Magn Mater 1999;194:132–139.[CrossRef]
  44. Lubbe AS, Bergemann C, Riess H, et al. Clinical experiences with magnetic drug targeting: a phase I study with 4'-epidoxorubicin in 14 patients with advanced solid tumors. Cancer Res 1996;56:4686–4693.[Abstract/Free Full Text]
  45. Lemke AJ, Senfft von Pilsach MI, Lubbe A, Bergemann C, Riess H, Felix R. MRI after magnetic drug targeting in patients with advanced solid malignant tumors. Eur Radiol 2004;14:1949–1955.[CrossRef][Medline]
  46. Bollinger A, Butti P, Barras JP, Trachsler H, Siegenthaler W. Red blood cell velocity in nailfold capillaries of man measured by a television microscopy technique. Microvasc Res 1974;7:61–72.[CrossRef][Medline]
  47. Iacob G, Rotariu O, Strachan NJ, Hafeli UO. Magnetizable needles and wires: modeling an efficient way to target magnetic microspheres in vivo. Biorheology 2004;41:599–612.[Medline]



This article has been cited by other articles:


Home page
 RadiologyHome page
C. Riviere, M.-S. Martina, Y. Tomita, C. Wilhelm, A. Tran Dinh, C. Menager, E. Pinard, S. Lesieur, F. Gazeau, and J. Seylaz
Magnetic Targeting of Nanometric Magnetic Fluid loaded Liposomes to Specific Brain Intravascular Areas: A Dynamic Imaging Study in Mice
Radiology, August 1, 2007; 244(2): 439 - 448.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2392042110v1
239/2/415    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fortin-Ripoche, J.-P.
Right arrow Articles by Clément, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fortin-Ripoche, J.-P.
Right arrow Articles by Clément, O.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE