Tel.: 49-6221-422569, Fax: +49-6221-422585
Cilengitide inhibits progression of experimental breast cancer bone metastases as imaged noninvasively using VCT, MRI and DCE-MRI in a longitudinal in vivo study
Article first published online: 20 JUL 2010
Copyright © 2010 UICC
International Journal of Cancer
Volume 128, Issue 10, pages 2453–2462, 15 May 2011
How to Cite
Bäuerle, T., Komljenovic, D., Merz, M., Berger, M. R., Goodman, S. L. and Semmler, W. (2011), Cilengitide inhibits progression of experimental breast cancer bone metastases as imaged noninvasively using VCT, MRI and DCE-MRI in a longitudinal in vivo study. Int. J. Cancer, 128: 2453–2462. doi: 10.1002/ijc.25563
- Issue published online: 25 MAR 2011
- Article first published online: 20 JUL 2010
- Accepted manuscript online: 20 JUL 2010 12:00AM EST
- Manuscript Accepted: 2 JUL 2010
- Manuscript Received: 10 MAR 2010
- Deutsche Forschungsgemeinschaft. Grant Number: SFB-TR23
- bone metastases;
- volumetric computed tomography;
- dynamic contrast-enhanced MRI
The aim of this study was to investigate the effect of inhibiting αvβ3/αvβ5 integrins by cilengitide in experimentally induced breast cancer bone metastases using noninvasive imaging techniques. For this purpose, nude rats bearing established breast cancer bone metastases were treated with cilengitide, a small molecule inhibitor of αvβ3 and αvβ5 integrins (75 mg/kg, five days per week; n = 12 rats) and compared to vehicle-treated control rats (n = 12). In a longitudinal study, conventional magnetic resonance imaging (MRI) and flat panel volumetric computed tomography were used to assess the volume of the soft tissue tumor and osteolysis, respectively, and dynamic contrast-enhanced (DCE-) MRI was performed to determine functional parameters of the tumor vasculature reflecting blood volume and blood vessel permeability. In rats treated with cilengitide, VCT and MRI showed that osteolytic lesions and the respective bone metastatic soft tissue tumors progressed more slowly than in vehicle-treated controls. DCE-MRI indicated a decrease in blood volume and an increase in vessel permeability and immunohistology revealed increased numbers of immature vessels in cilengitide-treated rats compared to vehicle controls. In conclusion, treatment of experimental breast cancer bone metastases with cilengitide resulted in pronounced antiresorptive and antitumor effects, suggesting that αvβ3/αvβ5 inhibition may be a promising therapeutic approach for bone metastases.
Bone metastases occur frequently in many human malignancies including breast, prostate and lung carcinoma. The stimulation of osteoclasts by tumor cells proliferating within the bone marrow is a feature of the pathogenesis of bone metastases, and both the tumor and the bone microenvironment must be considered when strategies for therapy of bone metastases are developed.1 Bisphosphonates are potent inhibitors of osteoclast function that have been used over the last decades to treat patients with bone metastases. However, they do not induce regression of bone metastases. This, together with the adverse effects associated with bisphosphonate therapy such as osteonecrosis of the jaw and renal toxicity, emphasize the urgent need for the development of novel therapies that can be applied alternatively and as combination partners to target bone metastases more effectively.
Integrins are a family of 24 transmembrane proteins that integrate extracellular and intracellular activities. Besides their role in promoting physical adhesion, integrin signaling can induce cell spreading, migration, survival, proliferation and differentiation.2 The αvβ3 integrin interacts with several extracellular matrix (ECM) proteins including vitronectin, fibronectin, osteopontin, bone sialoprotein (BSP) and fibrinogen.3, 4 It is strongly expressed on activated tumor endothelial cells, whereas on resting endothelial cells in nondiseased tissues, its expression is generally low.5–7 In the pathogenesis of bone metastases, osteoclasts too express αvβ3 integrin, and selective αvβ3 inhibitors have been shown to inhibit osteoclast-mediated bone resorption in experimental prostate carcinoma bone metastases.8 Furthermore, αvβ3 integrin overexpression on tumor cells stimulated metastasis to bone in experimental models.9, 10 The closely related integrin αvβ5 is also a vitronectin receptor involved in breast cancer cell migration and invasion, but is less studied in the pathogenesis of bone metastasis, although it is overexpressed by osteoclasts and a wide range of cancer cells.11, 12 Together with αvβ5, αvβ3 integrin recognizes the arginine-glycine-aspartic acid (RGD) peptide sequence of extracellular ligands.13 Cilengitide (EMD 121974) is a cyclic pentapeptide containing the sequence RGDf(N-Me)V with high affinity for αvβ3 and αvβ5, which inhibits αvβ3/αvβ5-dependent cellular processes.14–17 As cilengitide inhibits αvβ3 and αvβ5 integrin from human, bovine and rat origin, it can be appropriately used in both experimental and clinical studies.15, 16 In recent phase II trials for treatment of glioblastoma multiforme, cilengitide has shown promising results including indications of antitumor activity and a good safety profile.13, 18 Cilengitide has antiangiogenic activity in model systems, correlating with its inhibition of attachment, migration, sprouting, differentiation, and in the induction of anoikis in those endothelial angiogenic cells whose adhesion and survival is dependent on αvβ3/αvβ5.15, 19, 20 Nevertheless, targeting αv integrins for therapy remains contentious, and for some tumors growth is accelerated in mice lacking αvβ3 and αvβ5, whereas in others, tumor growth and angiogenesis are accelerated by cilengitide.21, 22
In this study, we have used noninvasive imaging techniques to examine the dynamics of metastatic lesion development under therapy with cilengitide. Computed tomography (CT) and magnetic resonance imaging (MRI) are currently used to determine the extent of the osteolysis and the respective soft tissue component (STC) of bone metastases. For in vivo imaging of angiogenesis in bone metastases, dynamic contrast-enhanced (DCE-) MRI allows assessment of functional parameters associated with blood volume and vessel permeability in these skeletal lesions.23 We recently introduced an in vivo model of experimental breast cancer bone metastasis in which angiogenesis, soft-tissue lesion size, and extent of osteolysis can be monitored simultaneously and longitudinally by volumetric computed tomography (VCT), morphologic MRI and DCE-MRI.23, 24 Here we have used this model to noninvasively assess the treatment effects of cilengitide inhibiting αvβ3 and αvβ5 integrins in breast cancer bone metastases.
Materials and Methods
Cell lines and culture conditions
The human estrogen-independent breast cancer cell line MDA-MB-231 was purchased from American Type Culture Collection. Cells were cultured routinely in RPMI-1640 (Invitrogen, Karlsruhe, Germany) and supplemented with 10% FCS (Sigma, Taufkirchen, Germany). All cultures were kept under controlled conditions (humidified atmosphere, 5% CO2 and 37°C) and passaged two to three times a week to keep them in logarithmic growth.
The integrin expression profile of MDA-MB-231 human breast cancer cells was characterized using flow cytometry. Surface integrin staining on live cells was performed as described with minor modifications.25 Briefly, cells were harvested, rinsed, suspended in PBS-BSA (containing divalent cations) and sequentially incubated with mouse anti-αvβ3 (LM60926), mouse anti-αvβ5 (P1F627; Millipore, Schwalbach, Germany) or mouse anti-αv (17E625) followed by staining with fluorescinated goat-anti-mouse IgG and propidium iodide (5 μg/μl). Incubations used 10 μg/μl primary antibody concentrations and were for 45 min on ice. Flow cytometry was performed on a FACScan instrument (Becton-Dickinson, Heidelberg, Germany), gating for viable cells, and collecting 10,000 events per staining. The mean fluorescence intensity of the integrin staining was normalized using the staining intensity of the second layer reagent as background.
Animal model and therapy application
Experiments performed in this study were approved by the Local Governmental Animal Ethics Committee. Nude rats (RNU strain) were obtained from Harlan-Winkelmann GmbH (Borchen, Germany) at the age of 6 weeks and housed in a specific pathogen-free environment in a mini barrier system of the central animal facility of the DKFZ. Animals were kept under controlled conditions (21 ± 2°C room temperature, 60% humidity and 12 hr light-dark rhythm) and offered autoclaved food and water ad libitum. Subconfluent MDA-MB-231 cells were harvested using 0.05% Trypsin-EDTA (Gibco®; Invitrogen, Karlsruhe, Germany) counted on a Neubauer's chamber and resuspended in RPMI-1640 to a final concentration of 105 cells in 200 μl. Rats were anesthetized using a mixture of nitrous oxide (1 l/min), oxygen (0.5 l/min) and isoflurane (1.5 vol %). Arterial branches of the right hind leg were dissected and 105 cells injected into the superficial epigastric artery as described previously.28 Bone metastases were established and observed exclusively in the femur, tibia and fibula of the right hind leg. After 30 days of cancer cell transplantation, rats (n = 24) were randomly divided into two groups, one group receiving the cyclic RGD-peptide inhibitor of αvβ3/αvβ5 integrins (cilengitide, EMD 12197414, 17, 29; Merck, Darmstadt, Germany) intraperitoneally five times per week in isotonic saline (75 mg/kg; n = 12 rats) and the other, sham-treated group, serving as a control (n = 12 rats). The observation period of all animals was 55 days, and no rat in the study died ahead of schedule.
In vivo imaging
After the inoculation of cancer cells, each rat was imaged at days 30, 35, 45, and 55 using (i) a flat-panel equipped volumetric computed tomography (Volume CT; Siemens, Germany) and (ii) a 1.5-T clinical magnetic resonance scanner (Symphony; Siemens, Erlangen, Germany) equipped with a home-built receive-transmit coil (cylindrical volume resonator with an inner diameter of 83 mm and a usable length of 120 mm). Prior to in vivo imaging with VCT and MRI, rats were anesthetized with nitrous oxide, oxygen and isoflurane as described above.
Volumetric computed tomography
VCT imaging was obtained using the following parameters: tube voltage 80 kV, tube current 50 mA, scan time 51 sec, rotation speed 10 sec, frames per second 120, matrix 512 × 512, and slice thickness 0.2 mm. Image reconstructions were performed using a modified FDK (Feldkamp Davis Kress) cone beam reconstruction algorithm (kernel H80a; Afra, Erlangen, Germany).
Magnetic resonance imaging
The T2-weighted imaging was performed using a turbo spin echo sequence (orientation axial, TR 3240 ms, TE 81 ms, matrix 152 × 256, FOV 90 × 53.4 mm2, slice thickness 1.5 mm, 3 averages and scan time 3 min 40 sec). For DCE-MRI, a saturation recovery turbo flash sequence through the largest diameter of the tumor (orientation axial, TR 373 ms, TE 1.86 ms, matrix 192 × 144, FOV 130 × 97.5 mm, slice thickness 5 mm, measurements 512, averages 1, and scan time 6 min 55 sec) was used. After 20 sec baseline, 0.1 mmol/kg Gd-DTPA (Magnevist; Bayer Schering Pharma, Berlin, Germany) was intravenously infused for a time period of 10 sec.
Unenhanced VCT images and MRI-acquired T2-weighted images were analyzed using the Medical Imaging Interaction Toolkit (Heidelberg, Germany) to determine volumes of osteolytic lesions (OL) and (STCs), respectively. DCE-MRI acquired data was analyzed using the Dynalab workstation (Mevis Research, Bremen, Germany) according to the two-compartment model of Brix to determine the parameters amplitude A and exchange rate constant kep, as described.23, 30 Briefly, the injected contrast media is distributed in both compartments (intravascular space and extravascular, interstitial space). The accumulation of contrast agent in these compartments over time is characterized by the amplitude A (associated with blood volume), whereas the exchange of contrast agent between the intravascular space and the interstitial space is characterized by the exchange rate constant kep (associated with vessel permeability). For determination of the respective values of the amplitude A and kep of bone metastases in our study, a region of interest was placed around the STC on color maps for A and kep, respectively, using the Dynalab workstation (Mevis Research, Bremen, Germany).
At the end of the observation period, lower limbs of each animal were amputated and muscular tissue removed. Bones with surrounding soft tissue tumors were stored in 70% ethanol and embedded in a methylmethacrylat-based compound (Technovit 9100 NEU, Heraeus Kulzer, Hanau, Germany) according to the instructions of the manufacturer. The 5-μm thick sections were cut (Microm HM340e microtome; Thermo Scientific, Waltham, MA), mounted on SuperFrost Plus microscope slides and dried overnight at 60°C. In addition freshly removed soft tissue tumors were embedded in optimum cutting temperature (OCT) compound (TissueTec, Sakura, Japan) and stored at −80°C. The 7-μm thick cryosections (obtained on a Leica CM 3050S) were thaw-mounted, fixed in methanol and acetone and washed in PBS. For immunostaining, the Technovit-embedded sections were incubated overnight at 4°C with primary antibodies in PBS containing 12% bovine serum albumin. The following primary antibodies were used: rabbit anti collagen IV polyclonal antibody (1:50; Progen Biotechnik GmbH, Heidelberg, Germany) and mouse anti smooth muscle actin (SMA) polyclonal antibody (1:400; Sigma Aldrich, Saint Louis, MO). After washing in PBS, sections were incubated with secondary antibodies for 1 hr at room temperature as follows: Texas Red dye-conjugated donkey anti-rabbit IgG (1:100; Jackson Immunoresearch, Suffolk, UK) and Cy™2-conjugated goat anti-mouse IgG (1:50; Jackson Immunoresearch, Suffolk, UK). Cryosections were incubated overnight at 4°C with the following antibodies: mouse anti-human integrin αvβ3 Alexa Fluor 488 conjugated monoclonal [LM609] antibody (1:100; Millipore GmbH, Schwalbach, Germany) and mouse monoclonal [P1F6] antibody to integrin αvβ5 (Phycoerythrin) (1:100; Abcam, Cambridge, UK). After a nuclear staining step with DAPI (4′,6-diamidino-2-phenylindole, Serva, Heidelberg, Germany) sections were mounted in Fluoromount G (Southern Biotech, USA). Sections were examined using a Leica microscope (DMRE Bensheim, Germany) equipped with a digital camera (F-view XS; Soft Imaging System, Münster, Germany). Mean positive area fractions of SMA and collagen IV (in percent) as well as mean vessel diameters (in μm) were determined from four representative animals of each group analyzing 10 fields of view chosen randomly from each rat using Analysis Software (cell; Olympus Soft Imaging Solutions, Münster, Germany). Immunostainings for CD 31 (endothelial cells) and collagen IV (basal lamina) on tumor vessels were seen to be strongly positively correlated in STCs of bone metastases (data not shown).
For light microscopical analysis, sections were stained with Mayer's hematoxylin (Carl Roth, Karlsruhe, Germany) and eosin (Merck, Darmstadt, Germany), mounted using Eukitt mounting medium (O. Kindler, Freiburg, Germany) and analyzed using a microscope (DM LB; Leica, Wetzlar, Germany) equipped with a digital camera (DFC 320; Leica, Wetzlar, Germany).
For each animal, volumes of the osteolysis and STC, amplitude A and exchange rate constant kep were plotted vs. time after tumor cell inoculation (due to technical reasons one animal of the control group could not be evaluated for the amplitude A and kep). Normalization of the data to the corresponding initial value at Day 30 for each animal was performed, and changes were expressed in percent. For statistical comparisons of data from noninvasive imaging and histological analysis, the respective values were compared between the control and treatment groups using the two-sided Wilcoxon test; p values < 0.05 were considered significant.
MDA-MB-231 human breast cancer cells express αvβ5 but only low levels of αvβ3 integrins in vitro
The entire population of MDA-MB-231 cells in vitro expressed αv integrins as detected by the pan alpha-v reagent 17E6 (Fig. 1a). They showed low-cell surface expression of αvβ3 integrins by flow cytometry using the standard defining antibodies in the literature (36% of the cells were gated; median intensity 3-fold background), whereas staining strongly for αvβ5 integrins (100% cells gated; median intensity 10-fold background) (Figs. 1b and 1c). MDA-MB-231 also expressed α2, α3, α5, α6 and β1, β4, but not α4 or β6 chains (data not shown). In situ immunohistochemistry showed that soft tissue tumors stained strongly and quite uniformly for αvβ5 but had only weak patches of staining for αvβ3 (Fig. 1d).
Treatment with cilengitide reduces the volume of OL and STCs in experimental bone metastases as assessed in vivo with VCT and MRI
Tumor bearing animals were randomly assigned to two groups before therapy was begun at Day 30. The mean relative volumes of the OL and the STCs of bone metastases (STC) increased continuously in untreated rats until the end of the observation time (Day 55 post tumor cell injection) compared to the initial values at Day 30 after cancer cell injection (Fig. 2a). Mean relative values of the OL volumes increased by 1.9-, 4.5- and 9.7-fold in the control group and by 1.5-, 2.4- and 3.5-fold in the treatment group (at Days 35, 45 and 55, respectively) when compared to initial values at Day 30 (Fig. 2a, Fig. 3a). Significant differences between the groups were found at Days 45 (p < 0.05) and 55 (p < 0.01) for the OL (Fig. 2a). The mean volume of STC had increased by 2.3-, 10.4- and 22.5-fold in controls at Days 35, 45 and 55, respectively (Fig. 2a). However, the increase in mean relative STC values in bone metastases of the treatment group increased only by 2.2-, 4.9- and 6.3-fold for the volume of STC compared to initial values (Fig. 2a, Fig. 3b). Significant differences between the control and on-therapy groups were recorded at Days 45 (p < 0.05) and 55 (p < 0.01; Fig. 2a) for the STC. In the treatment group, three rats (25%) showed new bone formation under therapy with cilengitide as imaged by VCT (Fig. 3c). This bone formation was confined to the OL, and no excessive increase in bone mass beyond the osteolysis was observed. Such a de novo bone formation further confirmed by histology did not occur in control animals.
Experimental breast cancer bone metastases treated with cilengitide reveal changes in DCE-MRI derived parameters for both, relative blood volume and vessel permeability
For the mean relative values of the DCE-MRI parameter amplitude A, a significant decrease was found in animals treated with the αvβ3/αvβ5 inhibitor at Days 45 (102% of initial value; p < 0.05) and 55 (93% of initial value; p < 0.05) as compared to controls (Day 45, 125% and Day 55, 105% of initial values) but not on Day 35 postinoculation (106% in controls vs. 97% in treated rats; p > 0.05) (Fig. 2b, Fig. 4a). DCE-MRI parameter exchange rate constant kep also revealed significant differences at Day 55 postinoculation with increased values in treated animals (72% of initial value; p < 0.05) compared to controls (40% of initial value) but not on Days 35 (controls, 86% and treated animals, 69%; p > 0.05) or 45 (controls, 63% and treated animals, 88%; p > 0.05) (Fig. 2b, Fig. 4b).
Histological analysis reveals new bone formation, decreased vessel diameter and reduced colocalization of SMA and collagen IV in blood vessels of animals after treatment with cilengitide when compared to untreated controls
In control rats, bone metastases contained tumor cells (representing the soft tissue tumor) within areas of bone resorption corresponding to VCT and MR imaging (Fig. 5a). After treatment with cilengitide, newly formed bone was confirmed on hematoxylin/eosin-stained sections (Fig. 5b) taken from the proximal tibia of the animal, as shown in Fig. 3c. Immunofluorescence analysis in control animals revealed irregular vessels with small diameters, indicated by collagen IV staining in the basal lamina of vessels, which were not colocalized with SMA, along with larger vessels showing collagen IV/SMA colocalization (Fig. 5c). After 4 weeks treatment with cilengitide essentially only small and mesh-like vessels were seen without clear co-localization of SMA and collagen IV (Fig. 5d). Quantification of the immunofluorescent analysis resulted in significantly decreased mean positive area fractions of SMA (p < 0.05) and significantly increased area fractions of collagen IV (p < 0.01) in treated animals as compared to controls (Fig. 6a). The ratio of SMA and collagen IV (treated rats: 0.60/3.32; control rats: 0.83/2.37) was significantly decreased in animals after 4 weeks treatment with cilengitide (p < 0.01), and the mean vessel diameter in cilengitide-treated bone metastases (6.6 μm) was significantly smaller than in control rats (8.8 μm, p < 0.01; Fig. 6b).
The aim of this study was to assess the effects of the αvβ3/αvβ5 integrin inhibitor cilengitide on breast cancer bone metastases in nude rats transplanted with human MDA-MB-231 breast cancer cells. We used the noninvasive imaging techniques VCT, morphological MRI and DCE-MRI to follow-up longitudinal progression. Our primary findings are that cilengitide treatment, begun a month after tumors have been allowed to implant into bone, decreases osteolysis of breast cancer metastases in nude rats and the volume of the soft tissue tumor components. Cilengitide increases intratumoral vascular permeability, reduces the apparent numbers of mature intratumoral vessels and unexpectedly causes an resumption of bone formation in a quarter of the animals under therapy.
We found a significant decrease in osteolysis using VCT during therapy with cilengitide in nude rats. Several studies have reported a decrease of bone resorption in breast cancer bone metastases after inhibition of the integrin αvβ3.9, 31, 32 However, these groups have used MDA-MB-231 cells engineered and cloned to overexpress αvβ3 or breast cancer cell lines such as MDA-MB-435 that strongly express this integrin. As the MDA-MB-231 cells we used only express low levels of αvβ3, the antiresorptive effect observed here may not have been primarily due to the inhibition of this integrin on tumor cells, but also of αvβ3 on osteoclasts and on the intratumoral vasculature and αvβ5 integrin on all three compartments.12, 33 In previous studies, osteoclasts that express high levels of the αvβ3 integrin, bind several RGD-containing ECM proteins including vitronectin, osteopontin and BSP.34 By these interactions, αvβ3 is involved in the regulation of osteoclast activity, and the inhibition of this integrin was found to reduce osteoclast-mediated bone resorption.35 Furthermore, as angiogenesis is required for initiation and maintenance of osteoclastic bone resorption, its inhibition by cilengitide might have contributed to the observed decrease of osteolysis we observed after cilengitide treatment.36 As cilengitide cross reacts with human and rat αv integrins, the observed effects in our study are because of the inhibition of αvβ3 and αvβ5 integrins on both, MDA-MB-231 and host cells in particular of the vascular and bone compartments. Which compartments are targeted to produce the effects we report here is under investigation.
Interestingly, three animals (25%) treated with cilengitide here showed an increase in bone matrix, i.e., new bone formation in the OLs, which was not seen in control animals. There are no known therapies in use today for patients suffering from bone metastases, where such an effect is seen. After treatment with bisphosphonates, a sclerotic rim around OLs is a common sign for treatment response indicating local bone mineralization, but new bone formation is not seen after this therapy.37 Both integrins, αvβ3 and αvβ5, are expressed by osteoblasts and are associated with osteoblast migration, adhesion and activity.38 We have previously shown in this model of breast cancer bone metastases that the inhibition of BSP also resulted in decreased bone resorption and new bone formation.28, 39 As BSP binds αvβ3 integrin, the inhibition of either factors, BSP or αvβ3, might result in osteoblastic bone formation via the same pathway.40 However, the exact mechanism inducing bone regrowth must still be elucidated.
Not only were there antiresorptive effects but also the respective STCs had a lower volume than in the control animals, indicating an antitumor effect of cilengitide. Cilengitide inhibits the growth of several experimental tumors including melanomas and glioblastomas.41, 42 Because of the high expression of αvβ5 and the low expression of αvβ3 of MDA-MB-231 cells, the antitumor effect we report here may be a consequence of directly inhibiting αvβ5 on the surface of the breast cancer cells, combined with the antiangiogenic effects of inhibiting αvβ3 and αvβ5 on the endothelia of tumor vessels.15 However, this hypothesis was based only on the integrin expression of MDA-MB-231 cells observed in our study and has to be verified experimentally in further studies. Chen et al.43 previously observed that MDA-MB-231 cells expressed αvβ3 and αvβ5 integrins at similar levels suggesting that treatment effects of cilengitide might vary depending on the expression pattern of the respective cell clone used.
Antiangiogenic effects of cilengitide have been described previously in vitro and in vivo.15, 19, 41, 44 In our study, cilengitide treatment of experimental breast cancer bone metastases resulted in a decrease of the amplitude A and an increase of the exchange rate constant kep as assessed by DCE-MRI. These results indicated a decrease in blood volume and an increase of vessel permeability in these skeletal lesions, compatible with an “antiangiogenic” effect. In experimental glioblastomas and melanomas, a decrease in tumor vascularization and tumor growth followed treatment with cilengitide.21, 29 It is generally assumed that the antiangiogenic activity of cilengitide and related inhibitors is due to the experimentally observable inhibition of sprouting and differentiation and the induction of anoikis of angiogenic endothelial cells relaying on αvβ3 and αvβ5 for adhesion and survival.15, 45 In our immunohistological analysis, we observed vessel remodeling after cilengitide treatment including significantly decreased mean vessel diameter and SMA/collagen IV ratio, indicating that smaller vessels lacking pericyte and smooth muscle cells occurred more frequently in these animals than in untreated controls. These results of vessel remodeling rather than complete regression of tumor vessels upon cilengitide treatment are in good agreement with the moderate changes of DCE-MRI parameters A and kep. Taken together, we conclude that cilengitide triggered a decrease in blood volume (assessed by the amplitude A) due to smaller and partly nonfunctional blood vessels and increased vessel permeability (assessed by the exchange rate constant kep) was observed due to the increased number of immature vessels that arose after treatment with cilengitide.
Increased vessel permeability as seen in our study was previously reported by Alghisi et al.,46 who reported VE-cadherin delocalization from cell-cell contact sites on cilengitide treatment leading to a loss of cellular contacts and an increase of endothelial monolayer permeability. In bone metastases, this effect might improve local drug delivery to these lesions when combining cilengitide with standard treatments such as bisphosphonates or chemotherapy. In comparison with bisphosphonates showing predominantly antiosteoclastic and chemotherapy exhibiting mainly cytotoxic effects in bone metastases, cilengitide shows antiresorptive, antitumor and antiangiogenic efficacy in our study. Because of the favorable safety profile of this drug and the alternative mechanism of action compared to currently used treatments, cilengitide emerges as a promising novel therapy for breast cancer metastasis to bone and could be validated either as a single agent or in combination with bisphosphonates and chemotherapy in further experimental and clinical studies. Cilengitide might also be a suitable combination partner for ionizing radiation in the treatment of skeletal lesions due to its previously reported radio sensitizing effects in various tumors including breast cancer.47–49 In some rodent tumor models, a lack of αvβ3 and αvβ5 integrins or an inhibition by low concentrations of cilengitide stimulate tumor growth.50, 51 This seems not to be the case in the breast tumor-to-bone model we report here. Whether one or other of these experimental contexts better reflects the response of human pathologies to αv integrin inhibitors, however, must remain to be proven by clinical trial.19
In conclusion, treatment of well-established experimental breast cancer bone metastases with cilengitide resulted in an inhibition of bone resorption and soft tissue tumor growth in these osseous lesions and partial regrowth of bone. Although further experimental and clinical studies are required, cilengitide is a possible option for breast cancer patients suffering from metastases to bone.
The authors thank Karin Leotta, Renate Bangert, Lisa Seyler and Catherine Eichhorn for excellent technical assistance.
- 28Characterization of a rat model with site-specific bone metastasis induced by MDA-MB-231 breast cancer cells and its application to the effects of an antibody against bone sialoprotein. Int J Cancer 2005; 115: 177–86., , , , , .