Sage Journals: Discover world-class research

Abstract

The purpose of this study was to validate an integrin α_vβ₃–targeted magnetic resonance contrast agent, PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂, for its ability to detect tumor angiogenesis and assess early response to antiangiogenic therapy using dynamic contrast–enhanced (DCE) magnetic resonance imaging (MRI). Integrin α_vβ₃–positive U87 cells and control groups were incubated with fluorescein-labeled cRGD-conjugated dendrimer, and the cellular attachment of the dendrimer was observed. DCE MRI was performed on mice bearing KB xenograft tumors using either PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ or PEG-G3-(Gd-DTPA)₆-(cRAD-DTPA)₂. DCE MRI was also performed 2 hours after anti–integrin α_vβ₃ monoclonal antibody treatment and after bevacizumab treatment on days 3 and 6t. Using DCE MRI, the 30-minute contrast washout percentage was significantly lower in the cRGD-conjugate injection groups. The enhancement patterns were different between the two contrast injection groups. In the antiangiogenic therapy groups, a rapid increase in 30-minute contrast washout percentage was observed in both the LM609 and bevacizumab treatment groups, and this occurred before there was an observable decrease in tumor size. The integrin α_vβ₃ targeting ability of PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ in vitro and in vivo was demonstrated. The 30-minute contrast washout percentage is a useful parameter for examining tumor angiogenesis and for the early assessment of antiangiogenic treatment response.

FOR THE GROWTH AND METASTASIS of solid tumors, angiogenesis is a necessary process.^1,2 Focusing on angiogenetic factors, antiangiogenic therapy in conjunction with chemotherapy has already proved to be an effective treatment for non–small cell lung cancer, metastatic colon cancer, and metastatic breast cancer.^3–5 However, because these therapies are expensive and have potential adverse effects, a need exists to screen for tumor angiogenesis before initiating antiangiogenic therapy. A noninvasive imaging method would be invaluable if it could detect the presence of tumor angiogenesis, and, furthermore, help identify an individual who will respond to future antiangiogenic therapy. Integrin α_vβ₃ is highly expressed on proliferating vascular endothelial cells and is considered to be a biomarker of tumor-induced angiogenesis; its expression level correlates well with tumor invasiveness and disease state, whereas it is essentially absent on mature quiescent cells.^6–9 The introduction of anti–integrin α_vβ₃ monoclonal antibody (MAb) is able to promote tumor regression by inducing apoptosis of angiogenic blood vessels.^10,11 Owing to its restricted expression in tumors, together with its critical role in tumor angiogenesis, integrin α_vβ₃ is an attractive target for tumor angiogenesis therapy.²

A tripeptide arginine-glycine-aspartic acid (RGD) sequence is present in extracellular proteins such as vitronectin, fibrinogen, thrombospondin, and osteopontin.¹² The RGD motif has been found to have a high affinity for integrin α_vβ₃.^7,12,13 Furthermore, cyclic RGD derivatives, which are based on the RGD tripeptide sequence, have a 100-fold better inhibition ability in relation to cell adhesion to vitronectin compared to the linear variant.^14,15 Thus, cyclic RGD can be used as a potential ligand to develop targeted probes for integrin α_vβ₃.

Researchers have previously labeled cyclic-RGD peptides with various radionuclides or conjugated cyclic-RGD peptides with near-infrared (NIR) fluorescence dyes; these have been used in various tumor xenograft models to evaluate tumor integrin α_vβ₃ expression by positron emission tomography (PET), single-photon emission computed tomography (SPECT), and NIR fluorescence camera.^16–25 However, the spatial resolutions of PET or SPECT are often not satisfactory, and radionuclides are often expensive and labor intensive to create. NIR imaging has an inevitable tissue penetration problem, and this restricts its clinical usefulness. Magnetic resonance imaging (MRI) is able to create cross-sectional images that have better spatial resolution, and this approach has no signal attenuation problem.²⁶ Researchers have previously conjugated RGD derivatives with various T₁ or T₂ contrast agents and used MRI to evaluate integrin α_vβ₃ expression using a number of animal models.^27–33 In these studies, the authors showed the presence of a higher contrast enhancement ratio at the tumor angiogenesis sites at specific time points after contrast injection. However, it is sometimes hard to obtain hemodynamic information from exactly the same area once the position of the mouse is changed. As a result, obtaining unbiased enhancement data from a specific portion of the tumor is difficult.

We previously reported the advantage of using dynamic contrast–enhanced (DCE) MRI to detect folate receptor (FR)-positive tumors in a mouse xenograft tumor model using a gadolinium (Gd)-loaded polyethylene glycol (PEG)-dendrimer-folate conjugate.³⁴ We demonstrated that a 30-minute contrast washout percentage (CWP) was a useful parameter when differentiating FR-positive tumors from FR-negative tumors. In this study, we conjugated cyclic RGD with the Gd-loaded PEG-dendrimer as an integrin α_vβ₃ targeting magnetic resonance (MR) contrast agent. Using DCE MRI, a Gd-loaded PEG-dendrimer-cRGD conjugate was used in a mouse xenograft tumor model to evaluate tumor angiogenesis; it was also used to detect suitability for early antiangiogenic treatment whereby the patient will respond to injection of anti–integrin α_vβ₃ MAb¹⁰ and anti–vascular endothelial cell growth factor (VEGF) MAb (bevacizumab).³⁵

Materials and Methods

Contrast Medium

A third-generation (G3) dendrimer with 16 hydroxyl groups was prepared as described previously.³⁴ Next, the hydroxyl groups present at the surface of the dendrimers were used for conjugation with diethylenetriaminepentaacetic acid (DTPA). Later, the DTPA terminals were conjugated with c(RGDfK) (Peptides International Inc., Louisville, KY), as well as complexed with Gd ions (Figure 1).

Figure 1.

Synthetic scheme used to make the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ dendrimer.

Preparation of the PEG-G3-(DTPA)_m

DTPA (0.28 g, 40 equivalent [eq]), hexafluorophosphate (HBTU; 0.27 g, 40 eq), and hydroxybenzotriazole (HOBt; 0.19 g, 80 eq) were dissolved in 20 mL dimethyl sulfoxide (DMSO) under nitrogen atmosphere. The solution was stirred at room temperature for 24 hours. PEG-G3-OH₁₆ (0.1 g, 1 eq, 0.0178 mmol) and 1,5-diazabicyclo[4.3.0]non-5-ene (DBN) (0.07 mL, 16 eq) were dissolved in 3 mL of DMSO and stirred for 10 minutes. The DTPA solution was added dropwise into the solution of PEG-G3-OH₁₆ and stirred at room temperature for 2 days. The solution was then dialyzed against deionized water. Free DTPA units were removed by ultrafiltration using a molecular weight cutoff 1000 membrane.

The number of DTPA groups was determined by the chelatometric titration method. A known weight of the dendrimer was dissolved in deionized water, and ammonium buffer (pH 10) was added followed by a drop Eriochrome Black-T indicator (Thermo Fisher Scientific, Waltham, MA). This mixture was titrated against 0.1 M calcium chloride solution until the blue color turned to a reddish orange color.

Preparation of the PEG-G3-(DTPA)_m-n-(cRGD-DTPA)_n

PEG-G3-(DTPA)_m (130 mg, 0.01625 mmol), HBTU (154 mg, 25 eq), and HOBt (110 mg, 50 eq) were dissolved in 20 mL DMSO and stirred for 1 hour at room temperature. Cyclic (RGDfK) (25 mg, 2.5 eq) and N,N-diisopropylethylamine (DIPEA) (21 mg, 10 eq) were dissolved in 30 mL DMSO. The c(RGDfK) solution was added dropwise into the solution of PEG-G3-(DTPA)_m and stirred at room temperature overnight under a nitrogen atmosphere. The solution was dialyzed against water to remove excess residues and filtered by spin column (Centricon, Millipore Corp., Billerica, MA) to remove low-molecular-weight molecules. In addition, c(RADfK) (Peptides International) was used to synthesize PEG-G3-(DTPA)_m-n-(cRAD-DTPA)_n as the control contrast agent, and the same protocol was used.

Preparation of the PEG-G3-(Gd-DTPA)_m-n-(cRGD-DTPA)_n

PEG-G3-(DTPA)_m-n-(cRGD-DTPA)_n (47 mg, 0.0053 mmol) was dissolved in 5 mL deionized water. GdCl3 (0.097 M, 0.273 mL) was added to the solution, and the pH was adjusted to 6.0 to 6.5 by 0.1 N NaOH and then stirred for 3 hours. The solution was filtered by spin column to remove free Gd ions and low–molecular-weight molecules and then lyophilized.

The absence of free Gd ions was confirmed using xylenol orange as the indicator at a pH of 5.8 (acetate buffer). The absence of uncomplexed DTPA units was confirmed by adopting the titration method used to determine the number of DTPA units present in the dendrimer. The number of Gd ions doped was determined experimentally using inductively coupled plasma-atomic emission spectroscopy (ICP-AES; S-35, Kontron, Germany). The complexes were filtered using a 0.45 μm filter and lyophilized. In addition, some cRGD-conjugated dendrimers were labeled with fluorescein isothiocyanate (FITC) for the cell binding experiments.

T₁ Relaxation Measurements

The Gd-loaded dendrimers (0.2–1 mmol) were evaluated for their capacity to alter the relaxation rate of water using a nuclear magnetic resonance (NMR) spectrometer (20 MHz, 0.47 T; MQ-20, Brucker, Germany) at 37°C with a standard pulse program of inversion recovery (range of repetition time: 0–300 ms).

Cell Experiments

U87 MG human glioblastoma cells (American Type Culture Collection [ATCC], Manassas, VA) were used as integrin α_vβ₃–positive cells and KB human epidermoid cells (ATCC) were used as integrin α_vβ₃–negative cells based on the literature.^36,37 Real-time polymerase chain reaction (RT-PCR) and immunohistochemistry (IHC) were performed to assess the integrin α_vβ₃ expression. Cellular attachment experiments using FITC-labeled cRGD-conjugated dendrimer were carried out on U87 MG, KB, and small interfering ribonucleic acid (siRNA)-treated U87 MG cells.

Integrin α_v RT-PCR

Total ribonucleic acid (RNA) was extracted from U87 MG and KB cells by a Qiagen RNeasy kit (Qiagen, Hilden, Germany), and 1 μg of extracted total RNA was subjected to a reverse transcription reaction using a high-capacity complementary deoxyribonucleic acid (cDNA) reverse transcription kit (Applied Biosystems, Foster City, CA). The cDNA from 20 ng of the total RNA was used as a template. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and integrin α_v messenger RNA were quantified by a sequence detection system instrument (RPISM 7000, Applied Biosystems) using the TagMan Gene Expression Assay (Applied Biosystems). The integrin α_v primers and probe were purchased from Applied Biosystems (Assay ID: Hs00233808_m1).

Immunohistochemistry

After fixation of the U87 MG cells using 100% ethanol, IHC was performed using a mouse antihuman integrin α_vβ₃ MAb (LM609, 1:50 dilution; Millipore Corp.), and this was developed with biotinylated goat antimouse secondary antibody (NEF823, 1:250 dilution; PerkinElmer, Waltham, MA). The staining procedure was performed with a modified avidin-biotin-peroxidase complex technique (PK-6100, Vector Laboratories, Burlingame, CA). The slides were visualized with the chromogen diaminobenzidine (DAB) and counterstained with hematoxylin (Vector Laboratories). Control sections were processed identically but incubated with nonspecific isotype immunoglobulin (Vector Laboratories) rather than LM609. These slides were also visualized with the chromogen DAB and counterstained with hematoxylin (Vector Laboratories).

For cryosectioning of the KB cell xenograft tumors, the tumors were dissected and cut in half, covered with Tissue-Tek (Sakura, Torrance, CA), and then frozen in liquid nitrogen vapor. Tumor sections were cut (5 μm in thickness) with a Microm HM525 microtome (Bio-Optica, Milano, Italy) and were methanol/acetone fixed. The sections were next blocked using an mouse on mouse immunodetection kit (Vector Laboratories) and incubated with monoclonal mouse anti-human integrin α_vβ₃ antibody (LM609, 1:50 dilution, Millipore Corp.) for 2 hours. This was followed by development using biotinylated goat antimouse secondary antibody (Nb720-55595, 1:250 dilution, Novus, Littleton, CO) for 30 minutes, and then a modified avidin-biotin-peroxidase complex technique was performed as described above. Control sections were processed identically but incubated with nonspecific isotype immunoglobulin (Vector Laboratories) rather than LM609. All slides were visualized with the chromogen DAB and counterstained with hematoxylin (Vector Laboratories).

Cellular Attachment Experiments

The U87 MG and KB cells (5 × 10⁴ cells) were cultured on a covered glass overnight. FITC-labeled cRGD-conjugated dendrimer (2 mM) was added to the culture medium for 2 hours, and then the cells were washed and fixed with 100% ethanol. 4′,6-Diamidino-2-phenylindole (DAPI) (D8417, Sigma-Aldrich, St. Louis, MO) was used for nuclear staining. The slices were analyzed using a fluorescence microscope (Axioplan, Zeiss, Germany). Blue and green channels were used for DAPI and FITC fluorescence detection. Axiovision software (Zeiss) was used for image acquisition.

In addition, free cRGD competition and integrin β₃ siRNA knockdown experiments were performed. Two hundred–fold free cRGD (400 mM) was added to the culture medium as a competitor for integrin α_vβ₃ before adding the FITC-labeled dendrimer (2 mM). In another setting, the U87 MG cells were first treated with integrin α_v-specific siRNA 48 hours prior to the cellular uptake study to suppress integrin α_v expression. Integrin α_v-specific siRNA is commercially available and was purchased from Ambion (Austin, TX) (AM_16708). Untreated U87 MG cells were used for the positive control. RT-PCR was used to validate the extent of the integrin α_v knockdown. Scrambled integrin α_v siRNA– (Ambion) treated U87 MG cells were used as the negative control. The integrin α_v siRNA–treated U87 MG cells, the scrambled integrin α_v siRNA–treated U87 MG cells, and the normal U87 MG cells were used for the FITC-labeled dendrimer cell binding studies.

Animal Preparation and Tumor Model

All animal studies were approved by the Institutional Animal Care Committee at our institution. Isoflurane inhalation was used for mice anesthesia. We delivered 1% for maintenance and 5% for induction with oxygen using a precision vaporizer (VIP3000, Midmark, Versailles, OH). Carbon dioxide inhalation was used for euthanasia. Eight-week-old nude male mice (National Laboratory Animal Center, Taipei, Taiwan), weighing 18 to 24 g (average weight 21.2 g), were handled in accordance with government guidelines.

To evaluate tumor-related integrin α_vβ₃ expression during tumor angiogenesis, integrin α_vβ₃–negative KB cells were used to develop a xenograft tumor model. To induce solid tumors, 1 × 10⁶ KB cells were injected subcutaneously into the right flank fat pads of 37 mice. Within 30 to 40 days after implantations, each mouse developed right flank tumors of 12 ± 4 mm in size.

Experimental Groups

The KB tumor–bearing mice were divided into five groups: the PEG-G3-(Gd-DTPA)_m-n-(cRGD-DTPA)_n injection group (n = 12), the PEG-G3-(Gd-DTPA)_m-n-(cRAD-DTPA)_n injection group (n = 12), the anti–integrin α_vβ₃ MAb (LM609, Millipore Corp.) treatment group (n = 3), the anti-VEGF MAb (bevacizumab, Roche, South San Francisco, CA) treatment group (n = 4), and the nontreatment group (n = 6).

In the PEG-G3-(Gd-DTPA)_m-n-(cRGD-DTPA)_n and PEG-G3-(Gd-DTPA)_m-n-(cRAD-DTPA)_n injection groups, efforts were made to inject both contrast agents into the same mouse to avoid intersubject biases. However, owing to anesthetic death, animal motion during examination, or venous injection failure, eventually only six mice received both cRGD-conjugated and cRAD-conjugated contrast medium injections for the DCE MRI study.

In the LM609 treatment group, pretreatment DCE MRI was performed 1 day before treatment, and then the mice received an intravenous LM609 injection via the retrobulbar venous plexus (25 μg/mouse) the next day. Finally, they received posttreatment DCE MRI 2 hours after LM609 injection.

In the bevacizumab treatment group, after performing the pretreatment DCE MRI study, the mice received an intraperitoneal injection of bevacizumab on the same day (5 mg/kg, intraperitoneal injection). The mice received posttreatment DCE MRI studies 3 and 6 days after bevacizumab treatment. A second dose of bevacizumab (5 mg/kg) was given intraperitoneally on the third day after the posttreatment DCE MRI. One mouse was sacrificed 7 days after the initiation of treatment, and the tumor was excised for anti–integrin α_vβ₃ IHC. The other three mice were kept, and the tumor volume was measured periodically. The tumor volume was calculated as follows: μ/6 × length × width × height.³⁸ Six mice did not receive antiangiogenic treatment (intraperitoneal administration of saline on the same day as the treatment groups). The tumor volumes of these mice were also measured periodically to obtain the normal growth curve of the KB cell xenograft tumors.

DCE Magnetic Resonance Imaging

MRI of the mice was performed using a 1.5 T superconducting system with a C4 surface coil. Axial DCE MRIs were then obtained with a section thickness of 2 mm and a 12 cm field of view through the centers of the bilateral flank tumors. T₁-weighted gradient echo sequences, with TR/echo time (TE) = 188/3.4, number of excitation = 1, flip angle of 80°, and a 179 × 256 acquisition matrix, were used. Forty dynamic images were obtained within 30 minutes from each of the mice. A bolus of PEG-G3-(Gd-DTPA)_m-n-(cRGD-DTPA)_n or PEG-G3-(Gd-DTPA)_m-n-(cRAD-DTPA)_n dendrimer (0.1 mmol per kilogram of body weight) was administered manually through a 31-gauge needle, which was placed in the retrobulbar venous sinus of the mice before the MRI study and connected to a 300 μL syringe through a polyethylene tube.

Data Analysis

Signal intensity (SI) values were measured in the operator-defined regions of interest (ROI). The ROI were placed by one investigator (W.-T.C.) with the aid of a cursor and graphic display device and encircled the entire tumor using an axial image through the bilateral tumor centers. SI was then measured for the KB tumors. The SI values derived from the ROI were plotted against time (0–30 minutes) as a time-intensity curve (TIC) by a software system (Extended MR WorkSpace, Philips, Best, Netherlands). The baseline value for signal intensity (SI_base) on a TIC was defined as the mean signal intensity from the first two images. The peak signal intensity (SI_peak) was defined as the peak enhancement value during the first pass of contrast medium. SI_30min was defined as the signal intensity measured 30 minutes after contrast injection. The contrast enhancement rise time (T_rise) was defined as the time between SI_base and SI_peak. The contrast washout time (T_washout) was defined as the time between SI_peak and SI_max.

The peak enhancement percentages [(SI_peak – SI_base)/SI_base × 100%], the 30-minute CWP [(SI_peak – SI_30min)/SI_base × 100%], the enhancement slope [(SI_peak – SI_base)/(SI_base × T_rise) × 100%], and the 30-minute contrast washout slope [(SI_peak – SI_30min)/(SI_base × T_washout) × 100%] for each ROI were calculated and compared by Student t-test.³⁴

Patterns of TICs

The TICs were classified into three types. Type A represents an initial rapidly rising slope followed by a second slowly rising phase (persistent enhancement curve). Type B represents a rapidly rising slope (wash-in) during the early phase, followed by a plateau after the peak enhancement. Type C represents a rapidly rising slope (wash-in), the same as the upright portion of the type B curve, followed by a washout phenomenon during the latter phase. Any change from −5 to 5% was considered to be a type B curve (plateau). More than 5% change was considered a type A curve. Any change to less than −5% was considered to be a type C curve.³⁹

Statistical Analysis

The results are presented as mean ± standard deviation. Differences in results acquired from the DCE MRI were compared using the Student t-test. The Fisher exact test was used to analyze the distribution of TIC patterns across the different groups. A value of p < .05 was considered to show significant difference. Statistical analysis was done with Stata 7 for Windows (Stata Corporation, College Station, TX).

Results

Contrast Medium

The synthesis of PEG-G3 dendrimer proceeded as previously described.³⁴ The formation of the G3 dendrimer was confirmed by ¹H NMR spectrometry and matrix-assisted laser desorption/ionization–time of flight (MALDI-TOF).

Synthesis of the PEG-G3-(DTPA)₈

The number of DTPA units attached to the G3 dendrimer was calculated from the volume of calcium chloride consumed. The average number of DTPA units present in each molecule of the cRGD-conjugated G₃ dendrimer was calculated to be eight.

Synthesis of the PEG-G3-(DTPA)₆-(cRGD-DTPA)₂

The number of cRGD units attached was calculated from the MALDI-TOF experiments by comparing the molecular weight of the reactant and the product. The average number of cRGD units present in each molecule of the PEG-G3-(DTPA)₈ dendrimer was calculated to be two.

Synthesis of the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂

Inversion recovery spectroscopy demonstrated that the DTPA ligands were complexed with Gd, which was identified by the disappearance of the C-O stretching vibration around 1,200 cm⁻¹ and the shift of the carbonyl stretch from 1,638 to 1,598 cm⁻¹. The number of Gd ions doped on each dendrimer was 5.5, as calculated by ICP-AES. The molecular weight of the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ dendrimer was about 10,235 kDa (see Figure 1).

T₁ Relaxation Measurements

The Gd-loaded cRGD-conjugated dendrimers were evaluated for their capacity to alter the relaxation rate of water using an NMR spectrometer (20 MHz, 0.47 T) with a standard pulse program of inversion recovery. The calculated r1 of the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ was 4.3 (mM.S)⁻¹ per Gd [25.6 (mM.S)⁻¹ per dendrimer]. The r2 was 5.7 (mM.S)⁻¹ per Gd [33.9 (mM.S)⁻¹ per dendrimer].

Cell Experiments

Integrin α_v RT-PCR and Integrin α_v siRNA U87 MG

The integrin α_v RT-PCR was carried out on the U87 MG and KB cells using GAPDH as an internal control and showed that the relative expression level of integrin α_v in U87 MG cells was 32-fold higher than that in KB cells. The integrin α_v siRNA knockdown experiment showed that the integrin α_v expression level, measured by integrin α_v RT-PCR, was suppressed to 15.5% of the integrin α_v messenger RNA expression level of the untreated U87 MG cells. A cell viability test showed that the integrin α_v siRNA-treated U87 MG cells had the same viability as nontreated U87 MG cells.

Immunohistochemistry

Integrin α_vβ₃ IHC analysis of the U87 MG cells showed positive brown staining on the U87 MG cell surface. No such staining was present on KB cells or on the nonspecific immunoglobulin control slide (data not shown). When the cryosections of the KB cell xenograft tumors were examined before treatment, the integrin α_vβ₃ IHC analysis showed positive brown staining along the tumor vasculature (Figure 2A) but not on the KB cells or on the nonspecific immunoglobulin control slide. After 7 days of bevacizumab treatment, the cryosections of the posttreatment KB cell xenograft tumors showed much less brown staining along the tumor vasculature after exactly the same treatment, when the post–bevacizumab treatment KB tumors (Figure 2B) were compared to the nontreatment sections.

Cellular Attachment Experiments

After 2 hours of incubation of FITC-labeled cRGD-conjugated dendrimer (2 mM) with U87 MG cells, strong green fluorescence spots were detectable at the cell surface of U87 MG cells by fluorescence microscopy (Axioplan, Zeiss) (Figure 3, A and B), but these were not present in the FITC-labeled cRAD-conjugated dendrimer group (data not shown). In the competition setting, 200-fold free cRGD (400 mM) was added to the culture medium before adding the FITC-labeled dendrimer; no green fluorescence was then detected on the cRGD presaturated U87 MG cells (Figure 3C). In another setting, when the integrin α_v siRNA–treated U87 MG cells were incubated with the FITC-labeled dendrimer (2 mM) for 2 hours, very little green fluorescence was detectable by fluorescence microscopy (Figure 3D), but green fluorescence could be seen in the control GAPDH siRNA–treated U87 MG cells. The KB cells showed no green fluorescence in each situation (data not shown).

Figure 3.

Cellular attachment experiments. A–B, Fluorescence microscopy of U87 MG cells after incubation with the FITC-labeled cRGD-conjugated dendrimer for 2 hours. The merged green and blue (DAPI) channel images show the FITC-labeled cRGD conjugates attached to the U87 MG cells (×20 and ×100 original magnification). C, Fluorescence microscopy of U87 cells after incubation with the FITC-labeled cRGD conjugates plus 200× free cRGD in culture medium for 2 hours. The merged image showed no green fluorescence signal attached to U87 MG cells. D, Fluorescence microscopy of U87 MG cells pretreated with integrin α_v siRNA 48 hours earlier then incubated with the FITC-labeled cRGD conjugates for 2 hours. The merged image shows no green fluorescence signal attached to the integrin α_v siRNA–pretreated U87 cell surface. The FITC-labeled cRAD conjugates attachment experiment also showed no FITC signal attached to the U87 MG cells (data not shown).

MRI and Data Analysis

Comparison of DCE MRI Parameters between the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ and PEG-G3-(Gd-DTPA)₆-(cRAD-DTPA)₂ Injection Groups

The enhancement percentage and enhancement slope of the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ injection group (n = 12) and the control groups (n = 12) showed no significant difference (enhancement percentage: 10.8 ± 4.4% vs 12.1 ± 4.9%; enhancement slope: 2.5 ± 1.1%/min vs 2.8 ± 1.3%/min). The 30-minute CWP and 30-minute contrast washout slope of the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ injection and control groups showed a significant difference (30-minute CWP: −5.5 ± 6.4% vs 7.3 ± 8%; contrast washout slope: −0.2 ± 0.3%/min vs 0.3 ± 0.3%/min, p < .001) (Figure 4A).

Figure 4.

A, The dot plot distribution of the 30-minute contrast washout percentage in the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ and PEG-G3-(Gd-DTPA)₆-(cRAD-DTPA)₂ injection groups. The 30-minute contrast washout percentage was −5.5 ± 6.4% in the cRGD-conjugated dendrimer injection group and 7.3 ± 8% in the cRAD-conjugated dendrimer injection group, p < .001. Six mice received both cRGD-conjugated and cRAD-conjugated contrast medium (solid lines). The 30-minute contrast washout percentage in these six pairs of data still shows a significantly lesser value in the cRGD-conjugated contrast medium injection group (−7.4 ± 8.8% vs 10.2 ± 10.8%, p < .05). B, The time-intensity curve (TIC) obtained from the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ injection group, which shows a persistent enhancement pattern (type A). C, The TIC obtained from the same mouse but using PEG-G3-(Gd-DTPA)₆-(cRAD-DTPA)₂, which shows a contrast washout pattern (type C).

The distribution of TIC types in the two groups is shown in Table 1. All of the type A curves were found in the cRGD-dendrimer injection group (Figure 4B). Eight (75%) of 12 type B curves were found in the cRGD-dendrimer injection group. All of the eight type C curves were found in the cRAD-dendrimer injection group (Figure 4C). There was a statistically significant difference in the distribution of the TIC patterns between the two groups (p < .001) by Fisher exact test.

Table 1.

Distribution of Time-Intensity Curve Types in the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ and PEG-G3-(Gd-DTPA)₆-(cRAD-DTPA)₂ Injection Groups

	Type
Injection Group	A	B	C	Total
PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂	4	8	0	12
PEG-G3-(Gd-DTPA)₆-(cRAD-DTPA)₂	0	4	8	12

Type A represents a persistent contrast enhancement pattern. Type B represents a plateau pattern after the peak contrast enhancement. Type C represents a contrast washout pattern.

Assessment of the Treatment Response of LM609

In the pre-LM609 treatment DCE MRI studies, the average 30-minute CWP of tumors was −6.8 ± 4% (n = 3), whereas the average 30-minute CWP of tumors was 3.1 ± 2.2% at 2 hours after LM609 MAb injection (Figure 5) (p < .05). The LM609-treated KB tumors showed a slower tumor growth rate than the nontreated KB tumors after LM609 treatment (see Figure 5C).

Figure 5.

A, The time-intensity curve (TIC) of a KB tumor before LM609 treatment. A persistent enhancement pattern (type A curve) was seen. B, The TIC of the same tumor 2 hours after LM609 treatment. The enhancement pattern has changed to a type C curve. A quick increase in the 30-minute contrast washout percentage was observed 2 hours after the initiation of LM609 treatment (Tx). C, The tumor growth curves of an LM609-treated mouse and an untreated mouse. The LM609-treated mouse shows a slower tumor growth rate than the nontreated mouse. DCE MRI = dynamic contrast–enhanced magnetic resonance imaging.

Assessment of the Treatment Response to Bevacizumab

In the prebevacizumab treatment DCE MRI studies, the average 30-minute CWP of tumors was −8 ± 4.2% (n = 4). The average 30-minute CWP of the tumors 3 days after bevacizumab treatment was significantly increased to be 1.5 ± 2.2% and reached 7.9 ± 6.6% at 6 days after initiation of the bevacizumab treatment (n = 4). A significant progressive increase in the average 30-minute CWP was noted for all mice (Figure 6) during the first 6 days after treatment. All of the bevacizumab-treated tumors had switched their enhancement pattern to type C at 6 days after the initiation of bevacizumab treatment. A tumor volume decrease was observed in all three bevacizumab-treated mice (except the one sacrificed for the IHC study) during the 40-day observation period, whereas all nontreated mice showed a progressive enlargement in tumor volume (see Figure 6B).

Figure 6.

Progression of the time-intensity curve (TIC) pattern from type A to type C in a KB tumor–bearing mouse after bevacizumab treatment. A, The TICs were obtained from the same mouse before and on days 3, 6, 20, and 40 after bevacizumab treatment using cRGD-conjugated Gd-loaded dendrimers. B, The tumor growth curves of the bevacizumab-treated mouse and two other untreated mice. Two mice received intraperitoneal normal saline injection when the other mouse received bevacizumab treatment (5 mg/kg) (day 0, arrowhead). The days on which posttreatment dynamic contrast–enhanced magnetic resonance imaging took place are indicated by arrows. Another dose of bevacizumab was given 3 days after day 0, while at the same time, the other two mice received normal saline injection. C, A progressive increase in the 30-minute contrast washout percentage can be seen over the first 6 days after the initiation of bevacizumab treatment (Tx) in all four mice.

Discussion

Owing to the differential upregulation of integrin α_vβ₃ in proliferating versus quiescent endothelial cells, integrin α_vβ₃ is frequently regarded as a neovascular biomarker and is also an attractive target for molecular imaging. Our results showed that a cRGD-conjugated Gd-loaded dendrimer can target integrin α_vβ₃–positive U87 MG human glioblastoma cells and that this attachment phenomenon can be suppressed by adding excessive free cRGD and is also diminished in the integrin α_v siRNA–treated U87 MG cells. In the DCE MRI study, the cRGD-conjugated Gd-loaded dendrimer demonstrated a persistent enhancement phenomenon compared to the control cRAD-conjugated contrast agent. The specificity of this integrin α_vβ₃–targeting contrast agent was confirmed by the anti–integrin α_vβ₃ MAb treatment study, which also indicated the potential role of this cRGD-conjugated contrast agent in the assessment of anti– integrin α_vβ₃–targeted therapy.^37,40 The 30-minute CWP revealed a hemodynamic change 2 hours after LM609 treatment and 3 days after bevacizumab treatment. Our data demonstrated that the cRGD-conjugated Gd-loaded dendrimer has a potential role in identifying tumor angiogenesis and in monitoring treatment response to antiangiogenic therapy.

The RGD motif has high affinity for the integrin family of cell adhesion molecules; this is true not only for integrin α_vβ₃ but also for integrin α₅β₁.¹² Integrin α₅β₁ is also overexpressed in tumor blood vessels in response to angiogenic factors but is not expressed in quiescent cells.^41–43 Schmieder and colleagues showed that a cRGD-conjugated nanoparticle can also target integrin α₅β_1.³³ Hence, the cRGD conjugate is a robust ligand for detecting tumor angiogenesis because it can target various angiogenesis-related integrin family members.³³ Kim and colleagues showed that bevacizumab treatment can inhibit integrin family expression, including integrin α₅β₁ and α_vβ₃.³⁵ Hence, cRGD conjugates can serve as a potential beacon for evaluation of the treatment response to bevacizumab therapy.

In the pilot study, we applied the conventional enhancement method to compare the enhancement ratio of the different groups at various time points (4 hours, 8 hours, 12 hours, etc.) after contrast medium injection, as other authors did in previous integrin α_vβ₃–targeting MR studies.^27–33 However, we found that it was difficult to image in exactly the same plane of the xenograft tumors once we had rearranged the mouse position. We also observed a fluctuated baseline MR signal intensity at various time points, which was caused by MR unit internal variability. This might cause measurement bias problems between series of MRI studies. In a clinical situation, it also might not be practical to follow the patient for such a long period. DCE MRI can acquire enhancement data from exactly the same portion of tumor during an examination and assess the semiquantitative pharmacokinetic characteristics of the contrast agent, such as the 30-minute CWP. We have also demonstrated the benefit of 30-minute CWP in terms of its ability to differentiate FR-positive tumors in our previous study using FR-targeting contrast medium.³⁴ In this study, we demonstrated that a cRGD-conjugated macromolecular MR contrast medium was more “sticky” to active endothelial cells during tumor angiogenesis and that this phenomenon was revealed by a negative value for the 30-minute CWP. This gave a useful baseline reference in terms of monitoring the tumor hemodynamic response to antiangiogenic therapy.

DCE MRI is a valuable imaging tool for tumor angiogenesis.^44–46 A two-compartment bidirectional exchange kinetic model has often been used in DCE MRI studies. However, we think that the two-compartment model is not suitable for receptor-targeted MRI. The ligand-conjugated contrast medium has a high affinity for target molecules and should remain longer within the target tissues. As a result, it is difficult to create and use a suitable model or formula to assess the hemodynamics of the receptor-targeting contrast agents.

All of the bevacizumab-treated KB tumors showed a progressive decrease in tumor size over the 40-day observation period even though only two doses of bevacizumab were given on days 0 and 3. Our results showed that the KB cell xenograft tumor model is a bevacizumab responder. However, some targeted cancer therapies are particularly effective in a specific subpopulation of patients. Clinicians therefore need to be able to predict the therapeutic benefit of a particular treatment at the patient level. Piessevaux and colleagues reported that tumor shrinkage at week 6 is a reliable and accessible tool for predicting the efficacy of cetuximab when treating chemorefractory metastatic colorectal cancer.⁴⁷ In this study, integrin α_vβ₃–targeted DCE MRI has been demonstrated as having the ability to identify in advance bevacizumab responders even before there is a decrease in tumor size. This finding could have a significant impact on the assessment of treatment responses to novel targeted cancer therapies in the future.

To obtain an unbiased comparison of the hemodynamic data between the different groups, we measured the entire cross-sectional area of each tumor. However, based on our observations, the vascularity of the peripheral zone of the xenograft tumors was higher than the central zone and there was a peripheral enhancement pattern in KB xenograft tumors. When we applied a smaller ROI to the peripheral enhanced regions of the KB tumors, we found a persistent contrast fill-in pattern that was greater in the cRGD-dendrimer injection group (data not shown). Although a peripheral enhancement pattern also was observed with the control cRAD-dendrimer group, their TIC also showed a contrast washout phenomenon.

During the synthesis of the conjugation of DTPA with the PEG-G3 dendrimer, we found that it was difficult to saturate all of the 16 hydroxyl groups on the dendrimer. The average number of DTPA units per dendrimer was six to eight over several attempts. We presumed that this might be due to steric hindrance. However, our dendrimer carried an average of six Gd-DTPA units per dendrimer to allow MR signal magnification. The r1 of our contrast agent was 25.6 (mM.S)⁻¹ per dendrimer, which is 8.8-fold higher than a single molecule of Gd-DTPA [2.9(mM.S)⁻¹]. Our results demonstrated the beneficial effect of increased T₁ relaxivity per molecule when using dendrimeric Gd chelates.⁴⁸

To avoid intersubject variation, we intended to use both cRGD-conjugated and cRAD-conjugated Gd-loaded contrast agents on the same mouse when we began our study. However, owing to anesthesia problems and venous injection failure, eventually only six mice received both cRGD-conjugated and cRAD-conjugated contrast medium during the DCE MRI study. The difference in 30-minute CWP in these six pairs of data still showed a significant difference between the cRGD- and cRAD-conjugated contrast medium injection groups. Such pairwise comparison provided stronger and nonbiased evidence that the cRGD-conjugated Gd-loaded dendrimer stayed significantly longer at the sites of tumor-related angiogenesis.

We demonstrated a decrease in integrin α_vβ₃ expression in the KB cell xenograft tumor model by IHC and also demonstrated that the 30-minute CWP gradually increased over the 6 days after bevacizumab treatment. Nonetheless, it is possible that this phenomenon might also be contributed to by a decrease in vessel permeability after bevacizumab treatment, which is well documented in the literature.⁴⁹ However, the change in 30-minute CPW remains a potential parameter that is able to define the baseline hemodynamic status before antiangiogenic therapy while allowing semiquantitative assessment of the hemodynamic changes in response to antiangiogenic therapy.

We observed one case in the cRGD conjugate injection group where a mild contrast washout phenomenon was present during the first 30 minutes (see Figure 4a, left row). We presumed that it might be caused by tumor-related arteriovenous shunting.⁵⁰ In such case, the blood flow velocity of tumor microvasculature will be exceptionally fast, and contrast washout phenomenon could be accelerated.

One of the drawbacks of this study is that we used a 1.5 T MR magnet for mouse tumor imaging. The imaging resolution (0.47 mm/pixel), signal to noise ratio, and contrast to noise ratio were not as good as with high magnetic MR units designed for small animals. However, the average tumor size was 1.1 cm in our study (range 0.8–1.5 cm). This range in tumor size is clearly revealed by a 1.5 T MR magnet using a small wrist surface coil. Another drawback of this project is that we did not perform quantitative correlation analysis between the xenograft tumor integrin α_vβ₃ expression level with the 30-minute CWP, but we have demonstrated a qualitative negative correlation between the tumor integrin α_vβ₃ expression level and the 30-minute CWP (see Figure 2). Another limitation of this study is that we did not show how the 30-minute CWP changes in nonresponders after antiangiogenic therapy. This was because it is hard to develop an appropriate nonresponse animal model.

Figure 2.

The integrin α_vβ₃ immunohistochemistry (IHC) of cryosections of the nontreated and bevacizumab-treated KB xenograft tumors. A, Positive brown staining was noted along and around tumor vasculature within the nontreated KB tumor section (×40 original magnification). B, When the integrin α_vβ₃ IHC of the bevacizumab-treated KB cell xenograft tumor was examined, the staining procedure and timing were exactly the same as for the nontreated sections and there was much less brown staining along tumor vessels compared to the nontreated tumor section (×40 original magnification). The 30-minute cRGD-conjugated dendrimer washout percentage from the pretreatment xenograft tumor (A) was −7.4%, and the 30-minute cRGD-conjugated dendrimer washout percentage from the posttreatment xenograft tumor (B) was 4.9%.

In this study, we have demonstrated not only that the use of cRGD-conjugated Gd-loaded dendrimer, based on a negative 30-minute CWP acquired by DCE MRI, is able to detect overexpression of integrin α_vβ₃ but also that this approach provides a valuable parameter that can be used to assess treatment response to antiangiogenic treatment; the latter assessment can be done before there is a decrease in tumor size, which is the previous approach to such assessment. In conclusion, DCE MRI using a cRGD-conjugated Gd-loaded dendrimer is a promising semiquantitative methodology for molecular imaging of integrin α_vβ₃ expression in vivo and has a potential and promising role in the early assessment of antiangiogenic therapy.

Footnotes

Acknowledgments

We thank Mr. Yu Yo Hsiau and Miss Yu Huey Jam for their diligent laboratory work.

Financial disclosure of authors: This research was supported by National Science Council grant 96-2321-B-532-001-MY3.

Financial disclosure of reviewers: None reported

References

Folkman

. What is the evidence that tumors are angiogenesis dependent? J Natl Cancer Inst 1990;82:4–6, doi:10.1093/jnci/82.1.4.

Folkman

. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995;1:27–31, doi:10.1038/nm0195-27.

Sandler

Gray

Perry

. Paclitaxel-carboplatin alone or with bevacizumab for non-small-cell lung cancer. N Engl J Med 2006;355:2542–50, doi:10.1056/NEJMoa061884.

Hurwitz

Fehrenbacher

Novotny

. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 2004;350:2335–42, doi:10.1056/NEJMoa032691.

Miller

Wang

Gralow

. Paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. N Engl J Med 2007;357:2666–76, doi:10.1056/NEJMoa072113.

Varner

. The role of vascular cell integrins alpha v beta 3 and alpha v beta 5 in angiogenesis. EXS 1997;79:361–90.

Brooks

. Role of integrins in angiogenesis. Eur J Cancer 1996;32A:2423–9, doi:10.1016/S0959-8049(96)00381-4.

Brooks

Clark

Cheresh

. Requirement of vascular integrin alpha v beta 3 for angiogenesis. Science 1994;264:569–71, doi:10.1126/science.7512751.

Razorenova

McCabe

. VEGF-integrin interplay controls tumor growth and vascularization. Proc Natl Acad Sci U S A 2005;102:7589–94, doi:10.1073/pnas.0502935102.

10.

Brooks

Montgomery

Rosenfeld

. Integrin alpha v beta 3 antagonists promote tumor regression by inducing apoptosis of angiogenic blood vessels. Cell 1994;79:1157–64, doi:10.1016/0092-8674(94)90007-8.

11.

Yamamoto

Maeshima

Kitayama

. Tumstatin peptide, an inhibitor of angiogenesis, prevents glomerular hypertrophy in the early stage of diabetic nephropathy. Diabetes 2004;53:1831–40, doi:10.2337/diabetes.53.7.1831.

12.

D'Souza

Ginsberg

Plow

. Arginyl-glycyl-aspartic acid (RGD): a cell adhesion motif. Trends Biochem Sci 1991;16:246–50, doi:10.1016/0968-0004(91)90096-E.

13.

Haubner

Finsinger

Kessler

. Stereoisomeric peptide libraries and peptidomimetics for designing selective inhibiotrs of the αvβ3 integrin for a new cancer therapy. Angew Chem Int Ed Engl 1997;36:1374–89, doi:10.1002/anie.199713741.

14.

Aumailley

Gurrath

Muller

. Arg-Gly-Asp constrained within cyclic pentapeptides. Strong and selective inhibitors of cell adhesion to vitronectin and laminin fragment P1. FEBS Lett 1991;291:50–4, doi:10.1016/0014-5793(91)81101-D.

15.

Gurrath

Muller

Kessler

. Conformation/activity studies of rationally designed potent anti-adhesive RGD peptides. Eur J Biochem 1992;210:911–21, doi:10.1111/j.1432-1033.1992.tb17495.x.

16.

Janssen

Oyen

Dijkgraaf

. Tumor targeting with radiolabeled alpha(v)beta(3) integrin binding peptides in a nude mouse model. Cancer Res 2002;62:6146–51.

17.

Cai

Sam Gambhir

Chen

. Multimodality tumor imaging targeting integrin alphavbeta3. Biotechniques 2005;39(6 Suppl): S14–25, doi:10.2144/000112091.

18.

Haubner

Wester

. Radiolabeled tracers for imaging of tumor angiogenesis and evaluation of anti-angiogenic therapies. Curr Pharm Des 2004;10:1439–55, doi:10.2174/1381612043384745.

19.

Cai

Chen

. Dual-function probe for PET and near-infrared fluorescence imaging of tumor vasculature. J Nucl Med 2007;48:1862–70, doi:10.2967/jnumed.107.043216.

20.

Chen

Conti

Moats

. In vivo near-infrared fluorescence imaging of integrin alphavbeta3 in brain tumor xenografts. Cancer Res 2004;64:8009–14, doi:10.1158/0008-5472.CAN-04-1956.

21.

Cheng

Xiong

. Near-infrared fluorescent RGD peptides for optical imaging of integrin alphavbeta3 expression in living mice. Bioconjug Chem 2005;16:1433–41, doi:10.1021/bc0501698.

22.

Haubner

Weber

Beer

. Noninvasive visualization of the activated alphavbeta3 integrin in cancer patients by positron emission tomography and [18F]galacto-RGD. PLoS Med 2005;2:e70, doi:10.1371/journal.pmed.0020070.

23.

Wang

. Dual optical and nuclear imaging in human melanoma xenografts using a single targeted imaging probe. Nucl Med Biol 2006;33:349–58, doi:10.1016/j.nucmedbio.2006.01.001.

24.

Cai

Cao

. (64)Cu-labeled tetrameric and octameric RGD peptides for small-animal PET of tumor alpha(v)beta(3) integrin expression. J Nucl Med 2007;48:1162–71, doi:10.2967/jnumed.107.039859.

25.

Sancey

Ardisson

Riou

. In vivo imaging of tumour angiogenesis in mice with the alpha(v)beta (3) integrin-targeted tracer 99mTc-RAFT-RGD. Eur J Nucl Med Mol Imaging 2007;34:2037–47, doi:10.1007/s00259-007-0497-z.

26.

Hollingworth

Todd

Bell

. The diagnostic and therapeutic impact of MRI: an observational multi-centre study. Clin Radiol 2000;55:825–31, doi:10.1053/crad.2000.0546.

27.

Winter

Caruthers

Kassner

. Molecular imaging of angiogenesis in nascent Vx-2 rabbit tumors using a novel alpha(nu)beta3-targeted nanoparticle and 1.5 tesla magnetic resonance imaging. Cancer Res 2003;63:5838–43.

28.

Schmieder

Winter

Caruthers

. Molecular MR imaging of melanoma angiogenesis with alphanubeta3-targeted paramagnetic nanoparticles. Magn Reson Med 2005;53:621–7, doi:10.1002/mrm.20391.

29.

Mulder

Strijkers

Habets

. MR molecular imaging and fluorescence microscopy for identification of activated tumor endothelium using a bimodal lipidic nanoparticle. FASEB J 2005;19:2008–10.

30.

Lee

Chen

. PET/MRI dual-modality tumor imaging using arginine-glycine-aspartic (RGD)-conjugated radiolabeled iron oxide nanoparticles. J Nucl Med 2008;49:1371–9, doi:10.2967/jnumed.108.051243.

31.

Huo

Zhang

. Gd-EDDA/HYNIC-RGD as an MR molecular probe imaging integrin alphanubeta3 receptor-expressed tumor-MR molecular imaging of angiogenesis. Eur J Radiol 2010;73:420–7, doi:10.1016/j.ejrad.2008.11.025.

32.

Boles

Schmieder

Koch

. MR angiogenesis imaging with Robo4- vs. alphaVbeta3-targeted nanoparticles in a B16/F10 mouse melanoma model. FASEB J 2010;24:4262–70, doi:10.1096/fj.10-157933.

33.

Schmieder

Caruthers

Zhang

. Three-dimensional MR mapping of angiogenesis with alpha5beta1(alpha nu beta3)-targeted theranostic nanoparticles in the MDA-MB-435 xenograft mouse model. FASEB J 2008;22:4179–89, doi:10.1096/fj.08-112060.

34.

Chen

Thirumalai

Shih

. Dynamic contrast-enhanced folate-receptor-targeted MR imaging using a Gd-loaded PEG-dendrimer-folate conjugate in a mouse xenograft tumor model. Mol Imaging Biol 2010;12:145–54, doi:10.1007/s11307-009-0248-6.

35.

Kim

Chung

Hong

. Bevacizumab application delays epithelial healing in rabbit cornea. Invest Ophthalmol Vis Sci 2009;50:4653–9, doi:10.1167/iovs.08-2805.

36.

Xiao

Yao

Peng

. Near-infrared optical imaging in glioblastoma xenograft with ligand-targeting alpha 3 integrin. Eur J Nucl Med Mol Imaging 2009;36:94–103, doi:10.1007/s00259-008-0920-0.

37.

Abdollahi

Griggs

Zieher

. Inhibition of alpha(v)beta3 integrin survival signaling enhances antiangiogenic and antitumor effects of radiotherapy. Clin Cancer Res 2005;11:6270–9, doi:10.1158/1078-0432.CCR-04-1223.

38.

Tomayko

Reynolds

. Determination of subcutaneous tumor size in athymic (nude) mice. Cancer Chemother Pharmacol 1989;24:148–54, doi:10.1007/BF00300234.

39.

El Khouli

Macura

Jacobs

. Dynamic contrast-enhanced MRI of the breast: quantitative method for kinetic curve type assessment. AJR Am J Roentgenol 2009;193:W295–300, doi:10.2214/AJR.09.2483.

40.

Kumar

. Integrin alpha v beta 3 as a therapeutic target for blocking tumor-induced angiogenesis. Curr Drug Targets 2003;4:123–31, doi:10.2174/1389450033346830.

41.

Boudreau

Varner

. The homeobox transcription factor Hox D3 promotes integrin alpha5beta1 expression and function during angiogenesis. J Biol Chem 2004;279:4862–8, doi:10.1074/jbc.M305190200.

42.

Bayless

Salazar

Davis

. RGD-dependent vacuolation and lumen formation observed during endothelial cell morphogenesis in three-dimensional fibrin matrices involves the alpha(v)beta(3) and alpha(5)beta(1) integrins. Am J Pathol 2000;156:1673–83, doi:10.1016/S0002-9440(10)65038-9.

43.

Bello

Francolini

Marthyn

. Alpha(v)beta3 and alpha(v)beta5 integrin expression in glioma periphery. Neurosurgery 2001;49:380–9; discussion 90.

44.

Jeong

Emerson

. Noninvasive evaluation of antiangiogenic effect in a mouse tumor model by DCE-MRI with Gd-DTPA cystamine copolymers. Mol Pharm 2010;7:41–8, doi:10.1021/mp900153f.

45.

Zwick

Strecker

Kiselev

. Assessment of vascular remodeling under antiangiogenic therapy using DCE-MRI and vessel size imaging. J Magn Reson Imaging 2009;29:1125–33, doi:10.1002/jmri.21710.

46.

Preda

Novikov

Moglich

. MRI monitoring of Avastin antiangiogenesis therapy using B22956/1, a new blood pool contrast agent, in an experimental model of human cancer. J Magn Reson Imaging 2004;20:865–73, doi:10.1002/jmri.20184.

47.

Piessevaux

Buyse

De Roock

. Radiological tumor size decrease at week 6 is a potent predictor of outcome in chemorefractory metastatic colorectal cancer treated with cetuximab (BOND trial). Ann Oncol 2009;20:1375–82, doi:10.1093/annonc/mdp011.

48.

Zhou

Luo

Tang

. Preparation of targeted Gd3+ contrast agent based on folate-receptor and its relaxation. Zhongguo Yaoxue Zazhi 2007;42:1640–4.

49.

Leach

Brindle

Evelhoch

. The assessment of antiangiogenic and antivascular therapies in early-stage clinical trials using magnetic resonance imaging: issues and recommendations. Br J Cancer 2005;92:1599–610, doi:10.1038/sj.bjc.6602550.

50.

Lau

Leung

. Arteriovenous shunts in patients with hepatic tumors. J Nucl Med 1997;38:1201–5.

Integrin α v β 3 –Targeted Dynamic Contrast–Enhanced Magnetic Resonance Imaging Using a Gadolinium-Loaded Polyethylene Gycol–Dendrimer–Cyclic RGD Conjugate to Evaluate Tumor Angiogenesis and to Assess Early Antiangiogenic Treatment Response in a Mouse Xenograft Tumor Model

Abstract

Materials and Methods

Contrast Medium

Preparation of the PEG-G3-(DTPA)m

Preparation of the PEG-G3-(DTPA)m-n-(cRGD-DTPA)n

Preparation of the PEG-G3-(Gd-DTPA)m-n-(cRGD-DTPA)n

T1 Relaxation Measurements

Cell Experiments

Integrin αv RT-PCR

Immunohistochemistry

Cellular Attachment Experiments

Animal Preparation and Tumor Model

Experimental Groups

DCE Magnetic Resonance Imaging

Data Analysis

Patterns of TICs

Statistical Analysis

Results

Contrast Medium

Synthesis of the PEG-G3-(DTPA)8

Synthesis of the PEG-G3-(DTPA)6-(cRGD-DTPA)2

Synthesis of the PEG-G3-(Gd-DTPA)6-(cRGD-DTPA)2

T1 Relaxation Measurements

Cell Experiments

Integrin αv RT-PCR and Integrin αv siRNA U87 MG

Immunohistochemistry

Cellular Attachment Experiments

MRI and Data Analysis

Comparison of DCE MRI Parameters between the PEG-G3-(Gd-DTPA)6-(cRGD-DTPA)2 and PEG-G3-(Gd-DTPA)6-(cRAD-DTPA)2 Injection Groups

Assessment of the Treatment Response of LM609

Assessment of the Treatment Response to Bevacizumab

Discussion

Footnotes

Acknowledgments

References

Preparation of the PEG-G3-(DTPA)_m

Preparation of the PEG-G3-(DTPA)_m-n-(cRGD-DTPA)_n

Preparation of the PEG-G3-(Gd-DTPA)_m-n-(cRGD-DTPA)_n

T₁ Relaxation Measurements

Integrin α_v RT-PCR

Synthesis of the PEG-G3-(DTPA)₈

Synthesis of the PEG-G3-(DTPA)₆-(cRGD-DTPA)₂

Synthesis of the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂

T₁ Relaxation Measurements

Integrin α_v RT-PCR and Integrin α_v siRNA U87 MG

Comparison of DCE MRI Parameters between the PEG-G3-(Gd-DTPA)₆-(cRGD-DTPA)₂ and PEG-G3-(Gd-DTPA)₆-(cRAD-DTPA)₂ Injection Groups