The Tyrosine Kinase Inhibitor, AZD2171, Inhibits Vascular Endothelial Growth Factor Receptor Signaling and Growth of Anaplastic

Cancer Therapy: Preclinical

TheTyrosineKinaseInhibitor,AZD2171,InhibitsVascularEndothelial Growth Factor Receptor Signaling and Growth of Anaplastic Thyroid Cancer in an Orthotopic Nude Mouse Model Fernando Gomez-Rivera,1AlfredoA. Santillan-Gomez,1Maher N.Younes,1Seungwon Kim,1 David Fooshee,1Mei Zhao,1Samar A. Jasser,1andJeffrey N. Myers1, 2

Abstract Purpose: Anaplastic thyroid cancer (ATC) is a locally aggressive type of thyroid tumor with high rate of distant metastases. With conventional treatment, the median survival ranges from 4 to 12 months; therefore, new treatment options are needed. AZD2171is a tyrosine kinase inhibitor of the vascular endothelial growth factor receptors (VEGFR) VEGFR-1,VEGFR-2, and VEGFR-3. The objective of the study is to determine whether AZD2171 can inhibit VEGFR-2 signaling and decrease tumor growth and prolong survival of ATC in an orthotopic nude mouse model. Experimental Design: We examined the effects of AZD2171 on phosphorylation of VEGFR-2, mitogen-activated protein kinase, and AKT in human umbilical vascular endothelial cells.Todeter- mine the antiproliferative and antiapoptotic effects of AZD2171, we did 3-(4,5-dimethylthiazol- 2-yl)-2,5-diphenyltetrazolium bromide and flow cytometry assays, respectively. We assessed the antitumor effects of AZD2171in a xenograft model of ATC using control, AZD2171,paclitaxel, and combination groups by measuring tumor size and survival. Results:Treatment with AZD2171led to dose-dependent inhibition ofVEGFR-2 phosphorylation and its downstream signaling in human umbilical vascular endothelial cells (IC50 for cell proliferation, 500 nmol/L). In the ATC cell lines DRO and ARO, IC50 was 7.5 Amol/L. AZD2171induced apoptosis in 50% of endothelial and ATC cells at 3 and 10 Amol/L concentrations, respectively. In vivo, AZD2171 led to a significant reduction in tumor size between control and AZD2171 (P =0.002)orAZD2171+paclitaxelgroup(P = 0.002) but not the paclitaxel alone group (P = 0.11). Survival was significantly higher among AZD2171 (P < 0.001) and combination groups (P < 0.001) compared with control. Conclusions: AZD2171effectively inhibits tumor growth and prolongs survival of ATC-bearing mice.The main effect of AZD2171is mediated through angiogenesis inhibition.

It is estimated that there will be 30,180 new cases of thyroid ranges from 4 to 12 months. Current treatment modalities for cancer in the United States in 2006. Thyroid cancer incidence is ATC include surgery, radiation therapy, and chemotherapy rapidly increasing and it is now the seventh most common (2, 3) The role of surgery is controversial and some authors cancer in women (1). Anaplastic thyroid cancer (ATC) accounts state that an aggressive approach does not provide additional for a small proportion of these cancers (3%); however, each benefit in most patients (4–9). year, it is responsible for more than half of the 1,200 deaths Radiation therapy may have an effect in delaying local from the disease. It is a locally aggressive type of tumor and has disease progression but it does not prolong survival (10–13), a high rate of distant metastases. With the conventional and for chemotherapy, a relatively recent phase II trial using treatment modalities, the median survival for this disease paclitaxel showed a 53% total response rate, without an improvement in survival (14). Given our poor ability to control ATC progression with conventional modalities, new paradigms are needed for treating this disease. Targeted Authors’ Affiliations: Departments of 1Head and Neck Surgery and 2Cancer Biology,The University of Texas M. D. Anderson Cancer Center, Houston,Texas molecular therapy is a promising approach to evaluate in the Received 11/1/06; revised 3/21/07; accepted 4/6/07. treatment of ATC patients. Our group has shown previously Grant support: M. D. Anderson Cancer Center Specialized Programs in Research that targeting epidermal growth factor receptor either with a Excellence in Head and Neck Cancer grant P50 CA097007A, NIH Cancer Center tyrosine kinase inhibitor (15) or with a monoclonal antibody support grant CA016672, AstraZeneca Pharmaceuticals, The University of Texas (16) is an effective approach for the treatment of ATC in an M. D. Anderson Cancer Center PANTHEONProgram. The costs of publication of this article were defrayed in part by the payment of page animal model. Moreover, we have shown that combined charges. This article must therefore be hereby marked advertisement in accordance inhibition of epidermal growth factor receptor and vascular with 18 U.S.C. Section 1734 solely to indicate this fact. endothelial growth factor receptor (VEGFR) with a dual Requests for reprints: Jeffrey N. Myers,The University of Texas M. D. Anderson tyrosine kinase inhibitor alone or combined with paclitaxel Cancer Center, 1515 Holcombe Boulevard, Houston,TX 77030. Phone: 713-745- 2667; Fax: 713-794-4662; E-mail: [email protected]. also produced significant cytostatic and cytotoxic effects on F 2007 American Association for Cancer Research. ATC cell lines in vitro and reduced tumor growth of s.c. ATC doi:10.1158/1078-0432.CCR-06-2636 xenografts in nude mice (17). The VEGF pathway is an ideal

www.aacrjournals.org 4519 Clin Cancer Res 2007;13(15) August 1,2007 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. Cancer Therapy: Preclinical therapeutic target because it is crucial for tumor growth and anti-MAPK and phospho-MAPK antibodies were used at a dilution of progression. Several approaches have been investigated, includ- 1:4,000; the anti-AKT and phospho-AKT antibodies were used at a ing agents that target either VEGF or its cell surface receptors dilution of 1:1,000. For secondary antibodies, we used goat anti-mouse (18) Molecules targeting VEGF include monoclonal antibodies secondary antibody at a dilution of 1:3,000 and a goat anti-rabbit antibody at a 1:3,000 dilution, both from Santa Cruz Biotechnology. and soluble receptor constructs. Receptor-targeted molecules Measurement of cytotoxicity. For the thyroid cancer cell lines, 2,000 include monoclonal antibodies and inhibitors of VEGFR cells were plated into wells of 96-well tissue culture plates. The cells tyrosine kinases. The small-molecule inhibitor AZD2171 is a were grown in RPMI 1640 supplemented with sodium pyruvate, highly potent angiogenesis inhibitor that selectively targets the essential amino acids, and 2% fetal bovine serum. For human umbilical VEGFRs VEGFR-1, VEGFR-2, and VEGFR-3 (19). These receptor vascular endothelial cell (HUVEC), 3,200 cells were plated into the tyrosine kinases, encoded by the FLT gene family, may fill 96-well plates and the cells were grown in EGM-2. After a 24-h distinct functions in regulating blood vessel growth and attachment period, the cells were refed with this medium or medium differentiation. VEGFR-2 exerts most of the functions of containing the VEGFR-2 inhibitor AZD2171. After incubation for VEGF in blood vessel endothelial cells, including survival, 3 days, the number of metabolically active cells was determined by nitric oxide production, vascular permeability, cell migration, MTT assay: the conversion of MTT to formazan was measured by a cell proliferation, and PGI production (20). VEGFR-2, like 96-well microtiter plate reader (MR-5000, Dynatech, Inc.) at an 2 absorbance of 570 nm. Growth inhibition was calculated with the many other receptors, induces proliferation through activation formula: cytostasis (%) = [1 - (A / B)] Â 100, where A is the absorbance of the classic extracellular signal-regulated kinase pathway of treated cells and B is that of control cells. [p42/44 mitogen-activated protein kinase (MAPK)], leading to Measurement of apoptosis. Thyroid cancer cells were plated at a gene transcription. VEGFR-2 also activates phosphatidylinositol density of 2 Â 105 cells/mL in six-well plates (Costar) and incubated at 3-kinase, which results in an increase in the lipid phosphati- 37jC for 24 h before treatment with AZD2171. HUVECs were also dylinositol (3,4,5) P3 and activation of several important plated at a density of 5,000 cells/cm2. Seventy-two hours later, the intracellular molecules, such as AKT (also denoted as protein extent of cell death was determined by propidium iodide staining of 5 kinase B) and the small GTP-binding protein Rac (21). VEGFR-3 hypodiploid DNA: 3 Â 10 cells were resuspended in Nicoletti buffer A is more abundant in lymphatic endothelial cells and its (50 g/mL; Sigma Chemical), 0.1% sodium citrate, 0.1% Triton X-100, activation induces proliferation, migration, and survival in and 1 mg/mL RNase (Roche) in PBS and then analyzed by fluorescence- activated cell sorting analysis (FACScan, Becton Dickinson). The those cells. VEGFR-1 regulates cell migration by modulating fraction of cells with sub-G1 DNA content was assessed using the Lysis actin reorganization, which is essential for cell motility. It has program (Becton Dickinson). The percentage of specific apoptosis was also been shown to negatively influence VEGFR-2 function in calculated by subtracting the percentage of spontaneous apoptosis of cells expressing chimeric VEGFR proteins. Previous studies have the relevant controls from the total percentage of apoptosis. shown that AZD2171 is able to reduce tumor size and prolong Western immunoblotting. HUVECs were incubated in six-well survival in animal models of colon, lung, prostate, and ovarian plates in serum-free medium without growth factors for 24 h. The cancer (19). The objective of the present project was to following day, they were incubated with AZD2171 for 90 min at determine if AZD2171 can inhibit VEGFR-2 signaling, the concentrations varying from 0.01 to 1 Amol/L before addition of VEGF main effector of the functions of VEGF in endothelial and/or (50 ng/mL) for 7 min. For thyroid carcinoma cell lines, the A tumor cells in vitro, and decrease tumor growth and prolong concentrations of AZD2171 varied from 0.1 to 10 mol/L. The cells survival in an orthotopic nude mouse model of ATC. were washed with PBS and lysis buffer was added [1% Triton X-100, 20 mmol/L Tris (pH 8.0), 137 mmol/L NaCl, 10% glycerol (v/v), 2 mmol/L EDTA, 1 mmol/L phenylmethylsulfonyl fluoride, 20 mmol/L Materials and Methods aprotinin-leupeptin-trypsin inhibitor, 2 mmol/L sodium orthovana- date]. The cells were scraped and centrifuged to remove insoluble Cell lines and culture conditions. The ATC cell lines ARO and DRO proteins. The samples were diluted in sample buffer [10% SDS, were used. The cells were grown in RPMI 1640 supplemented with 10% 0.5 mmol/L Tris-HCl (pH 6.8), 1 mol/L DTT, 10% (v/v) glycerol, 1% fetal bovine serum, L-glutamine, penicillin, sodium pyruvate, and bromphenol blue] and boiled. The proteins (40 Ag) were resolved by nonessential amino acids (Life Technologies, Inc.). Adherent monolay- PAGE and electrophoretically transferred onto polyvinylidene difluoride er cultures were maintained on plastic and incubated at 37jCin5% membranes. The membranes were blocked with 1% (w/v) bovine serum carbon dioxide and 95% air. The cultures were found to be free of albuminin0.1%Tween20(v/v)inTBS,probedwithmouse Mycoplasma species. Human umbilical vein endothelial cells were also monoclonal anti-VEGFR-2 antibody with 1% bovine serum albumin used (Cambrex) and maintained in EBM-2 medium. To split in 0.1% Tween 20 (v/v) in TBS, and incubated with peroxidase- endothelial cells, Clonetics HEPES-buffered saline solution was used conjugated sheep anti-mouse IgG antibody with 1% bovine serum for rinsing followed by trypsin/EDTA. After cells were released, trypsin albumin in 0.1% Tween 20 (v/v) in TBS. The blots were also probed with neutralizing solution was added to the plates. rabbit anti–phospho-VEGFR-2 antibody and phospho-MAPK antibody, Reagents. The VEGFR inhibitor AZD2171 was obtained from Astra- diluted with 1% bovine serum albumin in 0.1% Tween 20 (v/v) in TBS, Zeneca Pharmaceuticals and the National Cancer Institute, NIH and incubated with peroxidase-conjugated sheep anti-rabbit IgG (Bethesda, MD). For in vitro testing, the drug was prepared as antibody. Anti-h-actin was used at a 1:1,000 concentration. The blots 20 mmol/L stock solution by adding dimethyl sulfoxide as a vehicle were also probed with anti-AKT antibody and phospho-AKT antibody in and storing at -20jC. 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetra- a similar fashion. Protein bands were visualized using the Enhanced zolium bromide (MTT) and propidium iodide were purchased from Chemiluminescence Plus Western blotting detection system (Amersham Sigma-Aldrich Corp. Life Science). Materials. VEGFR-2 (KDR) antibody was purchased from Santa Animals. Eight- to 12-week-old male athymic nude mice were Cruz Biotechnology and phospho-KDR was from Biosource. Total and purchased from the National Cancer Institute and housed in a specific phosphorylated MAPK (p42 MAPK and phospho-p44/42 MAPK) and pathogen-free animal facility. All animal procedures were done in total and phospho-AKT antibodies were purchased from Cell Signaling accordance with a protocol approved by our Institutional Animal Care Technology, Inc. For Western blotting, the anti-KDR antibody was used and Use Committee. All mice were euthanized after 6 weeks by at a dilution of 1:500 and the phospho-KDR antibody at 1:1,000. The asphyxiation with carbon dioxide.

ClinCancerRes2007;13(15)August1,2007 4520 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. Inhibiting VEGFR-2 in ATC

Tumor xenograft generation for the treatment study with AZD2171 and paclitaxel. To produce tumors, DRO cells were harvested from subconfluent cultures by a brief 2-min exposure to 0.25% trypsin and 0.02% EDTA. Trypsinization was stopped with medium containing 10% fetal bovine serum. The cells were washed once and resuspended in serum-free medium. Five hundred thousand cells were suspended in 5 AL serum-free RPMI 1640 and injected orthotopically in the thyroid gland as described previously (22). I.p. injection of a ketamine/xylazine cocktail rendered the mice fully anesthetized. Then, the mice were restrained in dorsal decubitus and a midline cervical incision was done after 70% ethanol asepsis. Using a dissecting microscope, the submandibular glands were separated in the midline until the strap muscles were found. After retracting them to the left, the right thyroid lobe was exposed and a 25-AL Hamilton syringe connected to a 30-gauge needle was used to inject the cells. The wound was closed with surgical clips. All mice were examined the day after injection. After 5 days, time when tumors have already been established, all mice were weighed and randomized into four groups (n = 10 each): group one was treated with 1% (w/v) aqueous polysorbate 80; group 2 was treated with paclitaxel 200 Ag i.p. on day 1 of treatment and every week thereafter until the end of the study period; group 3 was treated with AZD2171 at a dose of 3 mg/kg via gastric gavage every day; and group 4 was treated with both AZD2171 and paclitaxel. Weight was measured twice weekly and mice were sacrificed at 3 weeks or earlier if they seemed morbid (hunched over, reluctant to move, difficulty with ambulation, ruffled haircoat, or loss of body weight of >20%). After the mice were euthanized, the xenografted tumors were obtained and measured with a caliper. The volume was calculated using the formula A Â B Â C Â p / 6, where A is the length of the longest aspect of the tumor, B is the length of the tumor perpendicular to A, and C is the depth of the tumor. For immunohistochemical and routine H&E staining, half of the tumor tissue was fixed in formalin and embedded in paraffin. The other half was embedded in ornithine carbamyl transferase compound (Miles, Inc.), rapidly frozen in liquid nitrogen, and stored at -70jC. Immunohistochemical detection of VEGFR-2, proliferating cell nuclear antigen, and CD31/platelet/endothelial cell adhesion molecule 1. Immu- nohistochemistry was done with the following antibodies: rabbit anti- VEGFR-2 (Santa Cruz Biotechnology) at a dilution of 1:100 and mouse anti–proliferating cell nuclear antigen (DAKO Corp.) at a dilution of 1:50. For VEGFR-2 staining, paraffin-embedded sections were first dewaxed in xylene. Excess xylene was removed by washing the slides in ethanol. After treating the tissue with pepsin for 20 min at 37jC, the slides were washed thrice with PBS. Endogenous blocking was done

Fig. 2. AZD2171inhibits proliferation of HUVEC but not ATC cell lines and induces apoptosis of HUVEC lines at high concentrations in vitro. A, a MTTassay was done in the presence and absence of AZD2171. A dose-dependent decrease in proliferation was seen in HUVECs treated with AZD2171.The concentration at which proliferation was inhibited by 50% (IC50)was0.5Amol/L. B, IC50 for ARO and DRO was found to be 7.5 Amol/L. A flow cytometry ^ based assay of propidium Fig. 1. AZD2171inhibits phosphorylation of VEGFR-2 in vitro.HUVECswere iodide ^ stained hypodiploid cells was used to measure the percentage of treated withVEGF stimulation for 7 min in the absence and presence of AZD2171. apoptotic cells after treatment with different concentrations of AZD2171for 48 h. This drug was able to completely inhibit VEGFR phosphorylation at a concentration C, AZD2171induced apoptosis of HUVEC starting at concentrations of 100 nmol/L of 0.1nmol/L. Activation of MAPK and AKTwas also decreased by treatment with a IC50 of 5 Amol/L. D, in contrast, concentrations of 10 and 20 Amol/L were with increasing concentrations of AZD2171. required to achieve the IC50 in ARO and DRO cell lines, respectively.

www.aacrjournals.org 4521 Clin Cancer Res 2007;13(15) August 1,2007 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. Cancer Therapy: Preclinical

Fig. 4. AZD2171improves the survival of athymic nude mice bearing orthotopic ATC xenografts. Kaplan-Meier survival curves showing the effects of AZD2171 and paclitaxel on the survival of mice in the orthotopic model of ATC. Group comparisons using the log-rank test. Statistically significant differences were seen for theAZD271and AZD2171 + paclitaxel groups when compared with the control.

fluorescence was minimized by covering the slides with 90% glycerol and 10% PBS. A positive reaction was visualized by incubating the slides with stable 3,3¶-diaminobenzidine for 10 to 20 min for identification of CD31/platelet/endothelial cell adhesion molecule 1. The sections were rinsed with distilled water, counterstained with Gill’s hematoxylin for 1 min, and mounted with Universal Mount (Research Genetics). Control samples that had not been exposed to primary antibody showed no specific staining. Staining for CD31/terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling and phospho-VEGFR-2 double immunofluorescence assays. Frozen tissues were used for terminal deoxynucleotidyl transferase–mediated dUTP nick end labeling (TUNEL) and CD31 double immunofluorescence assays. After being mounted on slides and fixation with acetone as described above, the frozen samples were washed thrice with PBS, incubated with protein-blocking solution containing 5% normal horse serum and 1% normal goat serum in in vivo Fig. 3. AZD2171inhibits orthotopic ATC xenograft growth . Tumors were PBS for 20 min at room temperature, and then incubated with a 1:400 measured in the three dimensions after mice were sacrificed. A, the two groups treated with AZD2171had a significant inhibition of tumor growth compared with dilution of rat anti-mouse CD31 monoclonal antibody (human cross- the control group. Columns, mean tumor volume; bars, SE. B, tumors in the control reactive) overnight at 4jC. After the samples were rinsed thrice with and paclitaxel alone groups had large tumors that deviated the trachea laterally PBS for 3 min each, the slides were incubated for 1 h at room and involved the right sternocleidomastoid muscle. Mice were sacrificed when they temperature in the dark with a 1:600 dilution of secondary goat anti-rat had lost >20% of their initial body weight in compliance with an Institutional Animal Care and Use Committee ^ approved protocol. antibody conjugated to Alexa Fluor 594 (red fluorescence). The samples were then washed thrice with PBS containing 0.1% Brij and once with PBS for 3 min. A similar protocol was used for phospho- with 3% H2O2 followed by protein blocking using 5% horse serum VEGFR-2 and CD31 double immunofluorescent staining. The antibody with 1% goat serum (protein-blocking solution). After washing the used for this purpose was rabbit anti–phospho-VEGFR-2 (Santa Cruz slides in PBS, the primary antibody (1:200 dilution) was added for Biotechnology). 18 h at 4jC. The slides were then washed with PBS thrice, blocked again TUNEL assay. A TUNEL assay was done using an apoptosis with protein-blocking solution for 1 h, and incubated with horseradish detection kit (Promega) with the following modifications: samples peroxidase–conjugated anti-rabbit antibody at 1:200 dilution for 1 h at were fixed with 4% paraformaldehyde (methanol free) for 10 min at room temperature. The slides were washed thrice in PBS and then room temperature, washed twice with PBS for 5 min, and incubated incubated with 3,3¶-diaminobenzidine for 10 min. After the excess 3,3¶-diaminobenzidine was washed off, counterstaining was done with Gill’s no. 3 hematoxylin. Table 1. Survival analysis by treatment group Frozen tissues were sectioned into 8- to 10-Am slices and used for detection of CD31/platelet/endothelial cell adhesion molecule 1. The Log-rank test P slices were mounted on positively charged Plus slides (Fisher Scientific) and air dried for 30 min; fixed in cold acetone (5 min), 1:1 acetone/ Control vs Paclitaxel 0.676 Control vs AZD2171 <0.001 chloroform (v/v; 5 min), and acetone (5 min); and then washed with Control vs Combination <0.001 PBS. Immunohistochemical procedures were done as described Paclitaxel vs AZD2171 0.001 previously with primary antibody diluted 1:400. Peroxidase-conjugated Paclitaxel vs Combination 0.003 secondary antibody was used for immunohistochemical analysis of AZD2171 vs Combination 0.253 CD31/platelet/endothelial cell adhesion molecule 1. Bleaching of

ClinCancerRes2007;13(15)August1,2007 4522 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. Inhibiting VEGFR-2 in ATC with 0.2% Triton X-100 for 15 min at room temperature. After two To quantify microvessel density (MVD), the areas containing the 5-min washes with PBS, the samples were incubated with equilibration highest number of tumor-associated blood vessels were identified by buffer for 10 min at room temperature. The equilibration buffer was scanning the tumor sections at low microscopic power (40Â). Vessels drained and a reaction buffer containing 44 AL equilibration buffer, that were completely stained with anti-CD31 antibodies were then 5 AL nucleotide mix, and 1 AL terminal deoxynucleotidyl transferase counted in 10 random 0.159-mm2 fields at Â100 magnification. (supplied in the kit) was added to the tissue sections and incubated Quantification of apoptotic endothelial cells was expressed as the in a humid atmosphere at 37jC for 1 h, avoiding exposure to light. average of the ratios of apoptotic endothelial cells to the total number The reaction was terminated by immersing the samples in 2Â SSC for of endothelial cells in 10 random 0.011-mm2 fields at Â400 15 min. Samples were then washed thrice for 5 min to remove magnification. unincorporated fluorescein-dUTP. Statistical methods. Best-fit curves were generated for the MTT and Quantification of microvessel density and apoptotic endothelial propidium iodide assays and used to determine the concentration at cells. For quantification analysis, five slides were prepared for each which 50% of the drug effect (IC50) was exhibited. Quantified results group, and two areas were selected on each slide. To quantify the of proliferating cell nuclear antigen, CD31, and tumor volume were expression in the TUNEL assays, the number of positively stained cells compared with Kruskal-Wallis and Wilcoxon rank-sum test, as and total cells were also counted in 10 random 0.159-mm2 fields of appropriate. Furthermore, a two-way ANOVA was used to test for the tumor area at Â100 magnification, and the percentages of positively effect of AZD2171 and paclitaxel as single agents and for interaction stained cells among the total number of cells were calculated and between compounds, which, when positively interacting, shows drug compared. synergy (23). Survival was analyzed with the Kaplan-Meier method. Immunofluorescence microscopy was done using a Leica DMLA Differences between the treatment and control groups were compared microscope (Leica Microsystems) equipped with an HBO 100 mercury with the log-rank test. A two-tailed P < 0.05 was considered significant. lamp and narrow band-pass filters to individually select for green, All statistical analyses were done using Stata 9 (Stata Corp.). red, and blue fluorescence (Chroma Technology Corp.). Images were captured using a cooled charged coupled device Hamamatsu 5810 Results camera (Hamamatsu Corp.) and ImagePro Plus 6.0 software (Media Cybernetics). Stained sections were examined in the same microscope AZD2171 inhibits phosphorylation of VEGFR-2 in vitro. We equipped with a three-chip charged coupled device color video camera (model DXC990, Sony Corp.). Photomontages were prepared using first examined the ability of AZD2171 to inhibit VEGFR-2 Photoshop software (Adobe Systems, Inc.). Endothelial cells were autophosphorylation (Fig. 1) in the endothelial cell line. identified by red fluorescence staining, and DNA fragmentation was HUVEC showed low levels of phospho-VEGFR-2 in serum-free detected by localized green fluorescence within the nuclei of apoptotic conditions when examined by Western blotting. However, cells. Photomontages were printed in a Sony digital color printer the addition of VEGF to the medium for 15 min resulted in (model UPD7000). the phosphorylation of both VEGFR-2 and AKT and a less

Fig. 5. AZD2171inhibitsVEGFR phosphorylation. Representative tumors. In the control group, CD31-positive cells colocalized with phospho-VEGFR-2 (pVEGFR2)andwere commonly seen in the periphery of the tumor. In the paclitaxel group, a similar pattern was observed. In contrast, staining for both markers was not as prominent for tumors in the AZD2171and AZD2171 + paclitaxel groups.

www.aacrjournals.org 4523 Clin Cancer Res 2007;13(15) August 1,2007 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. Cancer Therapy: Preclinical prominent decrease of MAPK. AZD2171 was able to completely AZD2171 inhibits proliferation of HUVEC but not ATC cell inhibit the VEGF-induced autophosphorylation of VEGFR-2 at lines in vitro. HUVEC and ATC cell lines ARO and DRO were 100 nmol/L (Fig. 1). At the same concentration, the phos- incubated with increasing concentrations (0-1 Amol/L for phorylation of MAPK and AKT were partially inhibited. HUVEC and 0-100 Amol/L for ATC cell lines) of AZD2171 in

Fig. 6. AZD2171induces endothelial cell apoptosis. A, representative tumors of the four groups.TUNEL staining was rarely seen in tumors in the control group, whereas CD31 staining was prominent. Paclitaxel-treated mice had staining forTUNEL but did not merge with the CD31-positive cells.The AZD2171and AZD2171 + paclitaxel groups had TUNEL staining that merged in some areas with CD31.Therefore, paclitaxel induced apoptosis of tumor cells, whereas its combination with AZD217 orAZD2171alone produced apoptosis of both tumor and tumor-associated endothelial cells. B, quantitative analysis ofTUNEL staining. *, mean count for paclitaxel (P =0.006)and combination groups (P = 0.003) compared with the control group was statistically significant.

ClinCancerRes2007;13(15)August1,2007 4524 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. Inhibiting VEGFR-2 in ATC

proliferation of 40% was achieved at 10 Amol/L of AZD2171 for both ATC cell lines (Fig. 2B). AZD2171 induces apoptosis of HUVEC lines at high concentrations in vitro. To examine if AZD2171 had proapoptotic effects, HUVEC and ATC cell lines (ARO and DRO) were treated with various concentrations of AZD2171 for 48 h. Propidium iodide staining showed that AZD2171 induced apoptosis of 20% of HUVECs at a concentration of 1 Amol/L and 50% at 5 Amol/L (Fig. 2C). For ARO and DRO cell lines, 50% induc- tion of apoptosis by AZD2171 only occurred at concentrations of 7.5 and 10 Amol/L, respectively (Fig. 2D). AZD2171 inhibits orthotopic ATC xenograft growth in vivo. We assessed the antitumor effects of AZD2171 in a xenograft model of ATC using four randomized groups of 10 animals each: control, AZD2171, paclitaxel, and combination groups by measuring tumor size and survival. All animals tolerated both AZD2171 and paclitaxel without significant weight loss in the four groups, and all animals in the control group remained alive up to day 20 of treatment. Paclitaxel alone did not significantly inhibit the growth of orthotopic ATC xenografts generated with the DRO cell line in nude mice when compared with control. In contrast, AZD2171 alone or in combination with paclitaxel inhibited the growth of tumors when compared with the control group (Fig. 3). At the end of the 3-week treatment period, the mice treated with paclitaxel, AZD2171, and AZD2171 plus paclitaxel showed 15%, 53%, and 46% decreases, respectively, in mean tumor volume compared with the control group. AZD2171 and AZD2171 plus paclitaxel combination produced statistically significant (P = 0.002) reductions in mean tumor volume when compared with the control group. In addition, a significant difference was found between the AZD2171 alone and paclitaxel alone groups (P < 0.001). The reduction in mean tumor volume by AZD2171 alone compared with Fig. 7. Treatment with AZD2171decreases MVD. A, representative sections of the combination with paclitaxel, however, was not statisti- tumors in the four groups. MVDwas measured as the count of CD31-positive cally significant (P = 0.13). The difference in mean tumor cells in 0.159-mm2 fields at Â100 magnification. B, quantitative analysis of MVD. Paclitaxel alone did not make a statistically significant change in the MVD. *, mean volume between the paclitaxel group and the control group count compared with control was statistically significant forAZD2171 (P =0.004) did not have statistical significance either (P = 0.11). The P and combination groups ( =0.002). two-way ANOVA model revealed that only AZD2171 was significantly associated with tumor volume (P < 0.001). EGM and RPMI 1640, respectively, complemented with 2% In contrast, no significant effect was seen with paclitaxel serum. After 72 h, the MTT assay showed that the proliferation (P = 0.43). of HUVEC was inhibited by AZD2171 in a dose-dependent AZD2171 improves the survival of athymic nude mice bearing manner (Fig. 2A). The concentration required to cause 50% orthotopic ATC xenografts. In the absence of any treatment, inhibition of cell growth (IC50) was determined to be 0.5 all of the control mice in the survival study succumbed to the Amol/L. In contrast, the concentrations required for inhibiting thyroid tumors by day 20 due to obstruction of the upper proliferation of ARO and DRO cell lines were significantly aerodigestive tract (Fig. 4). The median survival periods for higher at f7.5 Amol/L for both. The maximal inhibition of the control, paclitaxel, AZD2171, and combination groups

Table 2. Quantitative analysis of immunohistochemical staining of tumors in the four groups

Control Paclitaxel AZD2171 Combination P* MVDc 272 190 51 57 0.0005 MVD by fluorescence 327 194 50 22 0.002 TUNEL 305 2,678 604 5,942 0.0006

NOTE: Numbers are median values. *Statistical analysis by Kruskal-Wallis test. cMVD and TUNEL were measured as the number of counts of stained cells in 0.159-mm2 fields at Â100 magnification.

www.aacrjournals.org 4525 Clin Cancer Res 2007;13(15) August 1,2007 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. Cancer Therapy: Preclinical were 17, 16, 23, and 23 days, respectively. The differences in no study has shown a convincing improvement in survival. survival between the treatment groups were statistically Therefore, new treatment strategies are needed. One approach significant by log-rank test (P < 0.001). When comparing is to target angiogenesis; because tumors require access to groups, no statistically significant difference in survival blood vessels for growth and metastasis, inhibiting vessel was found between the control and paclitaxel-treated groups formation offers a way of reducing tumor size and hopefully (P = 0.67). However, a difference was evident for the groups prolonging survival for patients with ATC. In fact, MVD has that included AZD2171 (alone or in combination) compared been shown to relate to disease-free survival for patients with with the control and the paclitaxel alone groups (P < 0.001; papillary thyroid cancer (24) and to the spread of follicular Table 1). carcinoma (25). In addition, VEGF mRNA expression has been AZD2171 inhibits VEGFR phosphorylation, induces endothelial correlated with a lower 5-year overall survival of patients with cell apoptosis, and decreases MVD. A double-labeling immu- thyroid cancers (26). VEGF plays a key role in the neo- nohistochemical technique using antibodies against CD31/ vascularization of a tumor during cancer progression, as cells platelet/endothelial cell adhesion molecule 1 (red) and phospho- within the expanding mass of the tumor are frequently VEGFR-2 (green) showed by colocalization that endothelial cells deprived of oxygen because of their great distance from the within all the thyroid tumors expressed the activated VEGFR-2. nearest blood vessels. As a consequence, hypoxic regions begin The phosphorylation of the receptor on tumor-associated to form. This phenomenon induces the production of VEGF, endothelial cells was inhibited by AZD2171 alone and in which produces migration, proliferation, differentiation, and combination with paclitaxel but not in the control or the pac- survival of endothelial cells (27). litaxel alone groups (Fig. 5). AZD2171 is a potent small-molecule tyrosine kinase inhi- The CD31/TUNEL fluorescent double-labeling immunohis- bitor of VEGFR-2 that also has activity against VEGFR-1 and tochemical technique showed that tumors in the paclitaxel VEGFR-3. A previous study by Wedge et al. showed that the and the combination groups had the highest staining for drug is active against lung, prostate, colon, breast, and ovarian TUNEL (Fig. 6B). However, in the paclitaxel group, it was tumors in mouse models and that the effects are mainly evident that the TUNEL staining did not merge with CD31 mediated by its inhibition of angiogenesis. Phase II clinical staining and therefore represented apoptosis in the tumor trials are ongoing with this compound in patients with non– cells. There was a higher percentage of tumor-associated small cell lung carcinoma, squamous cell carcinoma of the apoptotic endothelial cells in mice treated with AZD2171 head and neck, and colorectal cancer.3 In the present study, we alone and in combination with paclitaxel than in the control tested the effect of AZD2171 in an orthotopic model of ATC. or paclitaxel alone groups. A representative section of endo- Confirming our in vitro observations, we showed that the main thelial cells undergoing apoptosis is shown in Fig. 6A. Tumors mechanism of action of AZD2171 in ATC is also targeting in the AZD2171 alone group also showed neoplastic cells with endothelial cells. This was further supported by in vivo data, positive staining for TUNEL. in which mice treated with AZD2171 had an increased To determine the effect of AZD2171 on tumor angiogenesis, apoptosis of endothelial cells and tumor cells and a decreased MVD was quantified by staining tumor sections for CD31/ MVD as measured by CD31 positive count. These findings are platelet/endothelial cell adhesion molecule 1. The difference in contrast to the effects of the cytotoxic agent paclitaxel, between the four groups was statistically significant (P < 0.001). which produced apoptosis in tumor but not endothelial cells. Tumors from the mice treated with AZD2171 alone and It is known that VEGFR-2 activates phosphatidylinositol AZD2171 in combination with paclitaxel, but not on the 3-kinase, which results in an increase in the lipid phospha- paclitaxel alone group, had significantly lower MVD values tidylinositol (3,4,5) P3 and activation of several important compared with the control group (P < 0.001; Fig. 7; Table 2). intracellular molecules, such as AKT. We showed decreased Thus, AZD2171, when given alone and in combination with phosphorylation of AKT at concentrations of 0.01 nmol/L in paclitaxel, leads to a decrease in VEGFR-2 phosphorylation, an HUVECs. Tumor cells in the animals treated with AZD2171 increase in endothelial cell apoptosis, and a concomitant also underwent apoptosis, likely through a hypoxia-induced decrease in tumor-associated microvessels. mechanism. Teicher et al. (28) postulated that combined administration Discussion of antiangiogenic and cytotoxic therapies would yield maximal benefit because such combinations would destroy two separate In our studies, the small-molecule tyrosine kinase inhibitor compartments of tumors: cancer cells and endothelial cells. AZD2171 was found to inhibit tumor growth and prolong This has proved true in the clinic for patients with colorectal animal survival in an animal murine model of ATC. The main cancer; bevacizumab in combination with irinotecan, fluoro- mechanism of the action of the compound was through uracil, and leucovorin resulted in a 5-month increase in inactivation of VEGFR-2 and downstream signaling in endo- median survival (29). Therefore, we sought to determine if the thelial cells. These effects correlated with the ability of combination with paclitaxel, a cytotoxic agent that produces AZD2171 to inhibit proliferation of endothelial cells and tubulin polymerization, may have a synergistic effect with induce their apoptosis both in vitro and in vivo. Interestingly, AZD2171. Interestingly, paclitaxel did not have a synergistic paclitaxel alone or when combined with AZD2171 did not add effect when combined with AZD2171 for inhibiting tumor size any significant benefit in terms of tumor growth or survival. and did not affect survival. A possible explanation for these ATC is one of the most aggressive solid tumors with a results is in the dose of AZD271; we used a dose of 3 mg/kg, median survival of less than 12 months. Current treatment modalities, including surgery, radiation therapy, and chemotherapy, can delay local tumor progression in some cases, but 3 http://www.astrazeneca.com/article/511390.aspx

Clin Cancer Res 2007;13(15) August 1, 2007 4526 www.aacrjournals.org Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. Inhibiting VEGFR-2 in ATC

which is half of the maximum dose proposed by Wedge et al. and prolongs survival better than when the known cytotoxic At this dose, the authors found inhibition of tumor volume of chemotherapeutic agents are added, it would be a major >50% in colon, lung, and prostate tumors. However, it is improvement in targeted therapy involving angiogenesis possible that in our model, this dose was high enough to inhibition. obscure any effect from the addition of paclitaxel. In addition, We conclude that AZD2171 inhibits tumor growth and it is possible that at this dose, angiogenesis inhibition from prolongs survival in a murine orthotopic model of ATC. The AZD2171 compromised the delivery of paclitaxel. The results effects of AZD2171 seem to be mediated mainly by inhibition could also be explained by the lack of activity of paclitaxel on of angiogenesis. Therefore, AZD2171 seems to be a promising DRO cell line, and the possibility of the cell line being agent to evaluate in the treatment of patients with ATC (30). resistant needs to be addressed. In the present study, we were able to show that AZD2171 alone was able to slow tumor Acknowledgments growth and increased animal survival without evidence of We thank Meirong Gu for her technical assistance with the immunohistochemical toxic effects in the mice. If an antiangiogenic agent like stainings and AstraZeneca Pharmaceuticals and the National Cancer Institute AZD2171 could be given at a dose that inhibits tumor growth Cancer Therapy Evaluation Program, NIH for providing AZD2171.

References 1. JemalA,SiegelR,WardE,MurrayT,etal.Cancersta- ation therapy. Int J Radiat Oncol Biol Phys 1993;26: L. VEGF receptor signallingin control of vascular tistics, 2006. CA Cancer J Clin 2006;56:106 ^ 30. 12 5 ^ 8 . function. Nat Rev Mol Cell Biol 2006;7:359 ^ 71. 2. Are C, Shaha AR. Anaplastic thyroid carcinoma: bio- 13. Pierie JP, Muzikansky A, Gaz RD, Faquin WC, Ott 22. Kim S, Park YW, Schiff BA, et al. An orthotopic logy, pathogenesis, prognostic factors, and treatment MJ. The effect of surgery and radiotherapy on out- model of anaplastic thyroid carcinoma in athymic nude approaches. Ann Surg Oncol 2006;13:453 ^ 64. come of anaplastic thyroid carcinoma. Ann Surg mice. Clin Cancer Res 2005;11:1713^ 21. 3. Pasieka JL. Anaplastic thyroid cancer. Curr Opin Oncol 2002;9:57 ^ 64. 23. Slinker BK. The statistics of synergism. J Mol Cell Oncol 2003;15:78 ^ 83. 14. Ain KB, Egorin MJ, DeSimone PA; Collaborative Cardiol 1998;30:723^31. 4. McIver B, Hay ID, Giuffrida DF, et al. Anaplastic Anaplastic Thyroid Cancer Health Intervention Trials 24. Ishiwata T, Iino Y,Takei H, Oyama T, Morishita Y. thyroid carcinoma: a 50-year experience at a single (CATCHIT) Group. Treatment of anaplastic thyroid Tumor angiogenesis as an independent prognostic in- institution. Surgery 2001;130:1028^ 34. carcinoma with paclitaxel: phase 2 trial using ninety- dicator in human papillary thyroid carcinoma. Oncol 5. Ain KB. Anaplastic thyroid carcinoma: a therapeutic six-hour infusion.Thyroid 2000;10:587 ^ 94. Rep1998;5:1343^8. challenge. Semin Surg Oncol 1999;16:64 ^ 9. 15. Schiff BA, McMurphy AB, Jasser SA, et al. Epi- 25. SegalK,ShpitzerT,FeinmesserM,SternY, 6. Shimaoka K, Schoenfeld DA, DeWys WD, Creech dermal growth factor receptor (EGFR) is overex- Feinmesser R. Angiogenesis in follicular tumors of RH, DeConti R. A randomized trial of doxorubicin pressed in anaplastic thyroid cancer, and the the thyroid. J Surg Oncol 1996;63:95^8. versus doxorubicin plus cisplatin in patients with EGFR inhibitor gefitinib inhibits the growth of ana- 26. Berger DP, Herbstritt L, Dengler WA, Marme D, advanced thyroid carcinoma. Cancer 1985;56: plastic thyroid cancer. Clin Cancer Res 2004;10: Mertelsmann R, Fiebig HH. Vascular endothelial 2155 ^ 60. 8594^602. growth factor (VEGF) mRNA expression in human 7. Ahuja S, Ernst H. Chemotherapyof thyroid carcinoma. 16. Kim S, Prichard CN, Younes MN, et al. Cetuximab tumor models of different histologies. Ann Oncol JEndocrinol Invest1987;10:303 ^ 10. and irinotecan interact synergistically to inhibit the 1995;6:817^ 25. 8. Nel CJ, van HeerdenJA, GoellnerJR, et al. Anaplastic growth of orthotopic anaplastic thyroid carcinoma 27. Shinkaruk S, Bayle M, Lain G, Deleris G. Vascular carcinoma of the thyroid: a clinicopathologic study of xenografts in nude mice. Clin Cancer Res 2006;12: endothelial cell growth factor (VEGF), an emerging 82 cases. Mayo Clin Proc 1985;60:51 ^ 8. 600^7. targetforcancerchemotherapy.CurrMedChem 9. KobayashiT, Asakawa H, Umeshita K, et al.Treatment 17. Kim S, Schiff BA,Yigitbasi OG, et al. Targeted mole- Anti-Canc Agents 2003;3:95 ^ 117. of 37 patients with anaplastic carcinoma of the thy- cular therapy of anaplastic thyroid carcinoma with 28. Teicher BA, Sotomayor EA, Huang ZD. Antiangio- roid. Head Neck 1996;18:36 ^ 41. AEE788. Mol CancerTher 2005;4:632^ 40. genic agents potentiate cytotoxic cancer therapies 10. De Crevoisier R, Baudin E, Bachelot A, et al. Com- 18. Donovan EA, Kummar S. Targeting VEGF in cancer against primary and metastatic disease. Cancer Res bined treatment of anaplastic thyroid carcinoma with therapy. Curr Probl Cancer 2006;30:7 ^ 32. 19 92;52 :6702 ^ 4. surgery, chemotherapy, and hyperfractionated accel- 19. Wedge SR, Kendrew J, Hennequin LF, et al. 29. HurwitzH,FehrenbacherL,NovotnyW,etal.Beva- erated external radiotherapy. Int J Radiat Oncol Biol AZD2171: a highly potent, orally bioavailable, vascular cizumab plus irinotecan, fluorouracil, and leucovorin Phys 2004;60:1137 ^ 43. endothelial growth factor receptor-2 tyrosine kinase for metastatic colorectal cancer. N Engl J Med 2004; 11. Junor EJ, Paul J, Reed NS. Anaplastic thyroid carci- inhibitor for the treatment of cancer. Cancer Res 350:2335^42. noma: 91patients treated by surgery and radiotherapy. 2005;65:4389 ^ 400. 30. BellettiB,FerraroP,ArraC,etal.Modulationofin Eur J Surg Oncol 1992;18:83 ^ 8. 20. Matsumoto T, Claesson-Welsh L. VEGF receptor vivo growth of thyroid tumor-derived cell lines by 12. Levendag PC, De Porre PM, van Putten WL. Ana- signal transduction. Sci STKE 2001;112:RE21. sense and antisense vascular endothelial growth fac- plastic carcinoma of the thyroid gland treated by radi- 21. Olsson AK, Dimberg A, Kreuger J, Claesson-Welsh tor gene. Oncogene 1999;18:4860 ^ 9.

www.aacrjournals.org 4527 Clin Cancer Res 2007;13(15) August 1,2007 Downloaded from clincancerres.aacrjournals.org on September 30, 2021. © 2007 American Association for Cancer Research. The Tyrosine Kinase Inhibitor, AZD2171, Inhibits Vascular Endothelial Growth Factor Receptor Signaling and Growth of Anaplastic Thyroid Cancer in an Orthotopic Nude Mouse Model

Fernando Gomez-Rivera, Alfredo A. Santillan-Gomez, Maher N. Younes, et al.

Clin Cancer Res 2007;13:4519-4527.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/13/15/4519

Cited articles This article cites 30 articles, 6 of which you can access for free at: http://clincancerres.aacrjournals.org/content/13/15/4519.full#ref-list-1

Citing articles This article has been cited by 4 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/13/15/4519.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/13/15/4519. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.