On Human Endometrial Cancer Growth in Athymic Mice1

Vol. 7, 4149–4155, December 2001 Clinical Cancer Research 4149

Effects of the New Selective Estrogen Receptor Modulator LY353381.HCl (Arzoxifene) on Human Endometrial Cancer Growth in Athymic Mice1

Rita C. Dardes,2 David Bentrem, viously) EnCa101 endometrial tumors were treated with Ruth M. O’Regan, Jennifer MacGregor Schafer, Tam or Arzox, we observed a stimulatory effect of both compounds in these models. and V. Craig Jordan3 Conclusions: The results indicate that Arzox may be a Robert H. Lurie Comprehensive Cancer Center [R. C. D., J. M. S., good first-line agent, but it may be ineffective at controlling V. C. J.], Division of Hematology Oncology [R. M. O. ], and Department of Surgery [D. B.], Northwestern University Medical the growth of endometrial cancer after exposure to Tam. School, Chicago, Illinois 60611 Our data suggest that Arzox stimulates endometrial tumor growth to at least the same extent as Tam, thereby suggest- ing a limited role as a second-line agent for the patient on ABSTRACT Tam who develops occult endometrial cancer. Purpose: Arzoxifene (Arzox) is a novel benzothiophene analogue with selective estrogen receptor modulator activity similar to raloxifene. Arzox is being developed as a treat- INTRODUCTION ment for breast cancer and has a predominantly antiestro- Endometrial carcinoma is the most common malignancy of genic effect on the rodent uterus. Our objectives were to the female genital tract, and it is estimated to account for verify whether the novel selective estrogen receptor modu- approximately 36,100 new cases and more than 6,500 deaths in lator, Arzox, can be a good first-line agent and also be 2000 in the United States (1). Several risk factors for the effective at controlling the growth of endometrial cancer development of endometrial carcinoma are related to an ex- after exposure to tamoxifen (Tam). tended exposure to unopposed estrogen action, such as in nul- Experimental Design: We compared the effects of Tam liparity, late menopause, obesity, diabetes mellitus, estrogen and Arzox on the growth of estrogen responsive ECC-1 replacement therapy, and Tam4 treatment. Tam has a low inci- endometrial cancer cells in vitro, and we determined their dence of serious side effects, however, it has a partial estrogen antitumor effects on ECC-1 and EnCa101 endometrial car- agonist effect in the human uterus (2). The National Surgical cinoma growth in athymic mice. Adjuvant Breast and Bowel Project-P1 study showed that Tam Results: We observed that estrogen receptor protein does not increase the risk of endometrial cancer in premeno- expression is down-regulated by Arzox to the same extent as pausal women, but it does increase the risk by 3- to 4-fold in raloxifene, whereas 4-hydroxytamoxifen, the active metab- postmenopausal women (3). It is important to stress, however, olite of Tam, does not affect estrogen receptor protein levels. that the stage and grade of endometrial cancers observed in Tam and Arzox inhibit the growth of Tam-naı¨ve ECC-1 postmenopausal women were the same as in the general popu- tumors in athymic mice. However when Tam-stimulated or lation (4). estrogen-stimulated (which had been treated with Tam pre- The mechanism of uterine carcinogenesis caused by Tam is unknown. Two possible mechanisms have been proposed: (a) conversion of the drug to electrophilic metabolites that damage cellular DNA; and (b) an estrogen-agonist action on the uterus, promoting endogenous lesions. In vitro studies with human Received 4/25/01; revised 8/7/01; accepted 9/19/01. tissues or cells do not show significant DNA adduct formation The costs of publication of this article were defrayed in part by the by Tam in endometrium. However, in vivo studies have dem- payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to onstrated contradictory results with DNA adduct formation in indicate this fact. human endometrium (5). It seems most likely that, in women 1 These studies were supported by NIH Specialized Programs of Re- treated with Tam, the increased incidence of endometrial cancer search Excellence in Breast Cancer 1P50 CA89018-01 (to V. C. J.), is related to its estrogenic effects promoting uterine cell prolif- Coordenaçaõ de Aperfeiçoamento de Pessoal de Nı´vel Superior Schol- arship-Research Program from the Ministry of Education-Brasilia, DF, eration and stimulating the growth of preexisting endometrial Brazil (to R. C. D.), a grant from Eli Lilly, Indianapolis, Indiana (to cancers, because 5 years is not enough time to complete the V. C. J.), the generous support of the Lynn Sage Breast Cancer Foun- carcinogenic process (6). In addition, women with both previous dation of Northwestern Memorial Hospital, and the Avon Products estrogen replacement therapy exposure and higher body mass Foundation. 2 Present address: Department of Gynecology, Federal University of Saõ Paulo, 740 Botucatu CEP: 04023-900, Saõ Paulo-Brazil. 3 To whom requests for reprints should be addressed, at Northwestern University Medical School, Robert H. Lurie Comprehensive Cancer Center, 303 East Chicago Avenue, 8258 Olson Pavilion, Chicago, IL 4 The abbreviations used are: Tam, tamoxifen; ER, estrogen receptor; 60611. Phone: (312) 908-5148; Fax: (312) 908-1372; E-mail: vcjordan@ Arzox, arzoxifene; Ral, raloxifene; SERM, selective estrogen receptor ␤ nwu.edu. modulator; E2,17 -estradiol; 4-OHT, 4-hydroxytamoxifen.

human endometrial carcinoma. The association between occult endometrial cancer and breast cancer is well known (18–20). If this paradigm is correct, then it is essential to evaluate the potential endometrial safety of a new SERM in the laboratory. We have used several different laboratory models of endometrial cancer to represent the application of Arzox treatment in the clinic. (a) endometrial cancer (ECC-1) never exposed to Tam; (b) an endometrial cancer initially exposed to chronic Tam

treatment but passaged in E2-treated athymic mice for 3 years (EnCa101-E2); and (c) EnCa101-Tam endometrial carcinoma, which is stimulated by Tam. These models represent: (a) patients never treated with Tam; (b) patients previously treated with Tam and now requiring a second-line endocrine therapy years after adjuvant Tam was stopped; and (c) patients treated with a second-line agent. Using our unique models of human endometrial carcinoma, we determine whether Arzox would be a viable agent for first- or second-line breast cancer in patients who develop occult endometrial cancer.

Fig. 1 Structures of SERMs referred to in the text. MATERIALS AND METHODS The ECC-1 human endometrial cancer cell line was de- rived from a primary well-differentiated, ER-positive endometrial adenocarcinoma (21). ECC-1 cells were maintained in index have the highest risk of Tam-associated endometrial can- flasks with phenol red containing DMEM supplemented with cer (6). 10% fetal bovine serum and 1% L-glutamine, 1% nonessential Tam (Fig. 1) is the endocrine adjuvant treatment of choice amino acids, 100 units/ml penicillin, and 100 mg/ml streptomy- for all ER-positive breast cancers (7). Not surprisingly, long- cin. Before experiments, cells were maintained in phenol red- term adjuvant Tam treatment does not control the recurrence of free DMEM with stripped fetal serum with the addition of 1% all ER-positive breast cancers, and drug resistance can develop L-glutamine, 1% nonessential amino acids, 100 unit/ml penicil- (8). In fact, some of these tumors seem to be stimulated by Tam lin, and 100 mg/ml streptomycin for at least 4 days. Cells were

(9, 10). Several molecular and cellular mechanisms have been grown at 37°C in a humidified 5% CO2 atmosphere. proposed as contributors to the acquisition of this resistant Growth Assay. The ECC-1 cells were harvested after phenotype (11, 12), including the selection of ER mutants, trypsinization and seeded into each well of 24-well plates at a alterations in the intracellular pharmacology and metabolism, concentration of 1.5 ϫ 104 cells (in 1 ml of estrogen-free ligand-independent ER-mediated transcription, and perturbation media). The growth assay was performed as described previ- of the interaction of ER with corepressors of transcription. ously (22), with doses of compounds ranging from 10Ϫ10 to Ϫ6 However, most of these Tam-stimulated tumors retain expres- 10 M. After 5 days of treatment, media was aspirated, 1 ml of sion of the ER (13) and respond to a second-line endocrine 0.1ϫ solution of hypotonic calcium, magnesium-free Hanks’ therapy (14). In addition to this problem, concerns about the balanced saline solution was added to each well, and cells were uterine safety of Tam (3, 4) have naturally provoked a search for chilled at 4°C for 10 min. The cells were sonicated, and 100 ␮l new first- or second-line agents that control the growth of both of the sonicated cell solution was transferred to an assay tube. A breast and endometrial carcinomas. standard DNA curve was completed using a stock concentration Arzox is a new benzothiophene analogue that is structur- of DNA (20 ␮g of calf thymus DNA/ml in hypotonic calcium, ally related to Ral (Fig. 1). Like Ral, Arzox is classified as a magnesium-free Hank’s balanced saline solution with 0.2 mg/ml SERM, based on its differential estrogenic/antiestrogenic effects BSA). The cell solution was incubated for at least1hin1.5ml in vivo on estrogen-target tissues (15). Arzox was designed of 1:200 solution of Hoechst dye (Polysciences, Inc.) and PES

to improve the bioavailability of Ral and provide sustained [2 M NaCl, 50 mM Na2HPO4,and1mM EDTA (pH 7.4)] per antiestrogenic blockade in the treatment of breast cancer. A assay tube. DNA concentrations were measured using a fluoro- preliminary clinical study of Arzox shows modest activity as meter. The activation wavelength was 356 nm, and the wave- a treatment for ER-positive metastatic breast cancer in post- length setting for reading samples was 458 nm. menopausal women (16). As Tam is used ubiquitously as an Western Blot Analysis. Western blotting was performed adjuvant in breast cancer treatment (7), any new SERM will as described previously (23). The ECC-1 medium was replaced Ϫ9 almost certainly be used after the failure of adjuvant Tam with fresh estrogen-free media containing E2 (10 M), 4-OHT Ϫ7 Ϫ7 Ϫ7 Ϫ7 therapy. In an earlier publication, we have reported that Arzox (10 M), Ral (10 M), Arzox (10 M), ICI182,780 (10 M), is effective as an antitumor agent in one model of Tam-stimu- or ethanol (control). The ECC-1 cells exposed to different lated breast cancer (T47D) but is cross-resistant in another compounds were harvested after 24 h treatment by scraping with (MCF-7) (17). a rubber policeman, in a PBS/EDTA solution. The cells were We now address the issue of Arzox and the growth of then pelleted by centrifugation, and the cell pellet was resus-

pended in protein extraction buffer [0.5% NP40, 2% glycerol, 1 mM DTT, 1 mM EDTA, 150 mM NaCl, 50 mM Tris (pH 7.4), 1 mM EGTA, 3 mM phenylmethanesulfonyl fluoride, 25 ␮g/ml leupeptin, 9 ␮g/ml aprotinin, 25 ␮g/ml trypsin inhibitor, and 25 ␮g/ml t-chymotrypsin]. Protein concentration from supernatant was measured using the Bio-Rad Protein Assay kit, and equal amounts of protein were run according to a standard Western blot protocol. The ER primary antibody used was AER311 (Neomarkers, Fremont, CA), and ␤-actin antibody AC-15 (Sig- ma Chemical Co., St. Louis, MO) was used to standardize loading. The appropriate secondary antibody conjugated with a horseradish peroxidase kit (Amersham Corp., Arlington Heights, IL) was used to visualize bands using an enhanced chemiluminescence visualization kit (Amersham Corp., Arling- ton Heights, IL). The membrane was exposed to Kodak X- OMAT film for 10 s. Athymic Mouse Model. The ECC-1 endometrial tumor was developed from 10 million ECC-1 cells/0.1 ml injected bilaterally into mammary fat pads of ovariectomized athymic mice 4–5 weeks of age (Harlan Sprague Dawley, Madison, WI).

E2 was administered viaa1cmsilastic capsule implanted s.c., which stimulated tumor growth (24). When ECC-1 tumors reached 1.0 cm2, a single tumor was harvested, and 1-mm3 tumor pieces were bitransplanted into the anterior axillary re- Fig. 2 The effect of E2, 4-OHT, and Arzox on the replication of gion of 30 athymic mice, which were treated as follows: 0.3 cm Ϫ9 ER-positive ECC-1 human endometrial cells in vitro.E2 (10 M) E2 capsules (eight mice); 1.5 mg of Arzox (eight mice); 1.5 mg stimulated ECC-1 cell proliferation compared with control (P Ͻ 0.001). ϩ of Arzox 0.3 cm E2 capsules (eight mice); or control (six In contrast, 4-OHT and Arzox did not stimulate ECC-1 cell proliferation Ϫ Ϫ mice). We also retransplated ECC-1 tumors into 40 athymic compared with control at concentrations of 10 9–10 6 M. However, at Ϫ10 mice divided into four groups of 10 and treated with 0.3 cm E low concentration (10 M), 4-OHT and Arzox slightly increased the 2 growth of ECC-1 cells (1-fold). We also demonstrated that 4-OHT and ϩ Ϫ capsules, 0.5 mg of Tam, 0.5 mg of Tam 0.3 cm E2 capsules, 9 Arzox inhibited the proliferation of E2-stimulated (10 M) ECC-1 cells or control. The EnCa101-Tam endometrial tumors used in these in a dose-dependent manner. In addition, Arzox seems more efficient experiments were developed in 1984 (25). EnCa101-E2 tumors than 4-OHT in blocking E2-stimulated growth of ECC-1 in vitro (EC50 ϫ Ϫ9 ϫ Ϫ8 were developed by passaging EnCa101-Tam tumors in mice Arzox, 2.4 10 M;EC50 4-OHT, 2.93 10 M). treated with E2 for 3 years (26). EnCa101-E2 endometrial tumors were transplanted into 60 athymic mice divided into six groups of 10 and treated with 0.5 mg of Tam, 0.5 mg of Tam ϩ ϩ 0.3 cm E2 capsules, 1.5 mg of Arzox, 1.5 mg of Arzox 0.3 cm administered by gavage at 0.5 or 1.5 mg, respectively, per

E2 capsules, 0.3 cm E2 capsules, or control. EnCa101-Tam mouse per day, 5 days/week. endometrial tumors were transplanted into 40 athymic mice and Statistical Analysis. Comparisons in mean tumor between the animal groups were analyzed by ANOVA at each divided into four groups of 10 then treated with 0.3 cm E2 capsules, 0.5 mg of Tam, 1.5 mg of Arzox, or control. Tumor week and then an unpaired Student’s t test. The two-tailed P of measurements were performed weekly using Vernier calipers. the last week of each experiment was reported using StatMost The cross-sectional area was calculated using the formula: 2.5 (Datamost Corp., Salt Lake City, UT). length ϫ width/4 ϫ␲. The Animal Care and Use Committee of Northwestern University, Chicago, IL, approved all procedures RESULTS involving animals. To examine the antiestrogenic activity of Arzox on endo-

Hormone and Drug Treatments. Silastic E2 capsules metrial cell proliferation in vitro, we used ER-positive ECC-1 (Sigma Chemical Co.) made as described previously (27) were endometrial cancer cells (Fig. 2), which are extremely sensitive implanted s.c. and replaced every 8 weeks. The 0.3 cm E2 to the stimulatory effects of E2: at physiological concentration Ϫ9 capsules produced a mean 83.8 pg/ml of serum E2, whereas (10 M), there was a 6-fold increase in DNA levels compared Ͻ 1.0 cm E2 capsules produced a mean 379.5 pg/ml of serum E2 with control (P 0.001). Arzox and 4-OHT slightly increased (26). Tam (Sigma Chemical Co., St. Louis, MO) or Arzox (1-fold) the proliferation of ECC-1 cells at a lower concentration Ϫ10 (generous gifts of Eli Lilly, Indianapolis, IN) were first dis- (10 M). However, neither 4-OHT nor Arzox increased the solved in ethanol and suspended in a solution of 90% carboxy- proliferation of endometrial cancer cells compared with control Ϫ9 Ϫ6 methylcellulose (1% carboxymethylcellulose in double-distilled at concentrations of 10 –10 M. We also demonstrated that Ϫ6 water) and 10% polyethylene glycol 400/Tween 80 (99.5% the antiestrogens, at high concentrations (10 M), completely Ͻ Ϫ9 polyethylene glycol 400 and 0.5% Tween 80). Ethanol was inhibited (P 0.001) the E2-stimulated (10 M) proliferation evaporated under nitrogen before use. Tam or Arzox were of ECC-1 cells (Fig. 2). Arzox was more potent in blocking the

␣ Fig. 3 Relative ER protein level in ECC-1 cells. E2 significantly decreased the level of ER␣, whereas 4-OHT had no effect. The pure antiestrogen, ICI182,780, Ral, and Arzox decreased the protein level of ER␣. The results demonstrate mean Ϯ SE on at least three determina- .P Ͻ 0.001, compared with control, by Student’s unpaired t test ,ء .tions

proliferative effects of E2 than 4-OHT at lower concentrations ϫ Ϫ9 ϫ Ϫ8 (EC50 Arzox, 2.4 10 M;EC50 4-OHT, 2.93 10 M). To characterize further the antiestrogenic activity of Arzox, we determined the effect on ER␣ protein levels in ECC-1 cells. It has been reported previously that estrogen decreases the transcription of ER␣ mRNA and protein levels in ECC-1 cells (22). Interestingly, 4-OHT slightly increases ER␣ expression in ECC-1 cells, but only at the protein level (22). It is widely accepted that the pure antiestrogen ICI182,780 destroys ER␣ protein (28). Ral decreases ER␣ expression in ECC-1 cells but

not as profoundly as E2 (Fig. 3). Because Arzox is a benzothiophene-like compound structurally similar to Ral, we examined ␣ the effects of Arzox on ER protein levels to confirm its Fig. 4 Inability of Arzox or Tam to stimulate growth in Tam-naı¨ve similarity with Ral. In Fig. 3, Western blot analysis shows ER␣ ECC-1 endometrial tumors in ovariectomized athymic mice (mean Ϯ ϭ protein levels in ECC-1 cells after treatment with E2, 4-OHT, SE). A, athymic mice (n 30) were bilaterally implanted with Tam- Ral, Arzox, and ICI182,780. Arzox decreased ER␣ protein naı¨ve ECC-1 tumors and treated with 0.3 cm E2 capsules, 1.5 mg of ϩ Arzox, 1.5 mg of Arzox 0.3 cm E2 capsules, or no treatment (control). levels similarly to Ral but to a lesser extent than ICI182,780 Arzox did not stimulate ECC-1 tumor growth and completely blocked (P Ͻ 0.01). These experiments were repeated three times. ϩ Ͻ the effects of E2 (Arzox E2 versus E2; P 0.001). Arzox was stopped Studies in vivo demonstrated the antiestrogenic action of and 1.0 cm E2 capsules were implanted to demonstrate the continued viability of the ECC-1 tumor cells at 14 weeks, (arrow). The mean Arzox. E2 maximally stimulated the growth of ECC-1 tumors tumor size of the groups with 1.0 cm E capsules after Arzox was transplanted into athymic mice (Fig. 4A). After 14 weeks of 2 stopped was 0.55 cm2 at 27 weeks (P ϭ 0.01; compared with control treatment, E2-stimulated growth was significantly greater than group). B, athymic mice (n ϭ 40) were bilaterally implanted with Ͻ all other treatments (P 0.001). In contrast, the 1.5 mg Arzox, Tam-naı¨ve ECC-1 tumors and treated with 0.3 cm E2 capsules, 0.5 mg ϩ ϩ or 1.5 mg Arzox 0.3 cm capsule E2-treated groups did not of Tam, 0.5 mg of Tam 0.3 cm E2 capsules, or no treatment (control). ϭ 2 grow after 14 weeks and were similar in size to the control E2 stimulated ECC-1 tumor growth; mean tumor 1.34 cm (E2 versus control; P Ͻ 0.001). Tam did not stimulate ECC-1 tumor growth and group. Mice were then implanted with 1.0 cm E2 capsules after ϩ Ͻ completely blocked the effects of E2 (Tam E2 versus control; P Arzox was stopped to determine whether the ECC-1 cells re- 0.001). mained viable. E2 significantly stimulated tumor growth compared with the control group (P ϭ 0.01), and the mean tumor size of was ϳ0.55 cm2 at 27 weeks. In the next experiment, we repeated this study with the ECC-1-transplanted tumors using EnCa101-E2 tumors were formed by transplanting

Tam treatment (Fig. 4B). Tam produced similar results when EnCa101-Tam tumors in mice treated with E2 (26). By the end compared with the previous experiment using Arzox (Fig. 4A). of week 12, the EnCa101-E2 tumors in the E2 group grew Ͻ E2 stimulated tumor growth significantly more than the other significantly more (P 0.001) than any of the other groups groups (P Ͻ 0.001). Tam did not exhibit agonist properties (Fig. 5). Additionally, 20 mice were treated with 0.5 mg of ϩ compared with control, and Tam completely blocked the effects Arzox (10 mice), or 0.5 mg of Arzox 0.3 cm E2 capsule (10

of postmenopausal E2. mice), but these data were omitted from the figure for clarity.

Fig. 5 Comparison of the growth of Tam and Arzox on EnCa101-E2 Fig. 6 Comparison of Tam and Arzox on EnCa101-Tam endometrial tumors. EnCa101-E2 endometrial tumors were bilaterally implanted into tumor growth (mean Ϯ SE). Forty ovariectomized athymic mice were 60 ovariectomized athymic mice (10/group) and treated as follows: bilaterally implanted with EnCa101-Tam endometrial tumors (10/group) ϩ 0.3 cm E2 capsules, 0.5 mg of Tam, 0.5 mg of Tam 0.3 cm E2 and treated for 9 weeks with 0.3 cm E2 capsules, 0.5 mg of Tam, 1.5 mg ϩ capsules, 1.5 mg of Arzox, 1.5 mg of Arzox 0.3 cm E2 capsule, or of Arzox, or control. E2, Tam, and Arzox stimulated EnCa101-Tam Ͻ control. E2 significantly stimulated tumor growth (mean size was 2.4 growth significantly more than control (P 0.001). Arzox was signif- cm2) compared with control and antiestrogen-treated groups. Both Tam icantly more agonistic (mean tumor ϭ 1.61 cm2) on the growth of and Arzox were partially agonistic compared with control (Tam versus EnCa101-Tam tumors than Tam; mean tumor ϭ 0.96 cm2 (P ϭ 0.023). control, P ϭ 0.039; Arzox versus control, P ϭ 0.001). But there was no significant difference between the antiestrogens. Both Tam and Arzox partially blocked postmenopausal estrogen, and there was no significant difference between the antiestrogens. DISCUSSION Tam increases the incidence of endometrial carcinoma in postmenopausal patients (4), possibly because it acts as a partial estrogen agonist. However, Ral seems to be less estrogenic than The low dose of Arzox (0.5 mg) also stimulated the growth of Tam on the rodent uterus (29), and Ral does not increase human EnCa101-E2 endometrial tumors to the same extent as 0.5 mg of endometrial thickness (30). No increase in the incidence of Tam and 1.5 mg of Arzox (data not shown). Interestingly, both endometrial cancer has yet been reported with Ral (31). The the 0.5 mg Tam and 1.5 mg Arzox treatments stimulated tumor incidence of endometrial cancer in the postmenopausal popula- ϳ growth compared with control (P Ͻ 0.005), but there was no tion is low ( 1/1000 women/year), so large study populations difference in tumor growth rates with either treatment (Fig. 5). are required to obtain accurate estimates of increased risk for However, when the mice were treated with 0.3 cm E capsules any new SERM. The increased incidence of human endometrial 2 cancer with Tam treatment (32) was predicted in the laboratory in addition to Tam and Arzox compounds, the tumors grew with the EnCa101 endometrial cancer model (33). We believe faster than with either antiestrogen alone (P ϭ 0.034, 0.5 mg of this model has the potential to provide valuable information Tam versus Tam ϩ E ; P ϭ 0.063, 1.5 mg of Arzox versus 2 about Ral derivative, Arzox, and endometrial cancer risk. Arzox ϩ E ). Tam and Arzox (at both doses) have similar 2 Recently we reported that Ral (0.5 mg/day or 1.5 mg/day agonist properties in EnCa101-E2, and both antiestrogens par- p.o.), stimulates the growth of Tam-stimulated MCF-7 breast Ͻ tially block the effects of E2 (P 0.001 compared with E2), but tumors and EnCa101-Tam endometrial tumors in vivo (34). there is no difference between the two antiestrogens. These These results suggest that if Ral is used for osteoporosis pre- results demonstrate that Arzox (at both doses) is cross-resistant vention after 5 years of Tam, it could result in the growth of with Tam in this tumor model. occult breast or endometrial cancers. Additionally, Ral seems to EnCa101-Tam endometrial tumors grow with either E2 and be cross-resistant with Tam in our short-term Tam-resistant Tam (26). In Fig. 6, E2 and Tam stimulated EnCa101-Tam breast cancer model (34), supporting the clinical finding that it growth, but there was no significant difference between the is not of value in the treatment of advanced Tam-refractory groups. By 9 weeks, the 1.5 mg-Arzox tumors had grown to a breast cancer (35). Previously, we have also observed that greater (P ϭ 0.023) extent than the 0.5 mg Tam group (Fig. 6). Arzox is cross-resistant with Tam in MCF-7-Tam breast tumors, In addition, 40 mice divided into four groups of 10 were treated and Arzox can only partially block estrogen-stimulated growth ϩ ϩ with 0.5 mg of Tam 0.3 cm E2 capsule, 1.5 mg of Arzox of T47D breast tumors in athymic mice (17). ϩ 0.3 cm E2 capsule, 0.5 mg of Arzox, or 0.5 mg of Arzox Although Ral is used to prevent and treat osteoporosis (30),

0.3 cm E2 capsule; but these data were omitted from the figure it has only 2% bioavailability (36), which compromises the for clarity. Low-dose Arzox (0.5 mg) stimulated the growth of proposed use of Ral as a breast cancer treatment. Arzox is a EnCa101-Tam tumors to the same extent as 0.5 mg Tam (data SERM designed to have an extended bioavailability compared not shown), and the combination of SERM with E2 stimulated with Ral (15) and is currently being developed for breast cancer the growth of EnCa 101-Tam tumors. treatment (37). Arzox is less estrogenic than Tam on rodent

uterus (15). Because Arzox may be used in early-stage breast after Tam failure in breast cancer patients. The mechanism of cancer, it is important to evaluate potential uterine side effects SERM-stimulated growth pathway in endometrium is not clear,

and document the effects on endometrial cancer growth in the but a recent report (42) demonstrates that Arzox, E2, and laboratory to predict the likelihood of Arzox-associated endo- Tam can activate uterine insulin-like growth factor I signaling metrial cancers. We have used three model systems of endome- pathways. trial cancer in vivo to simulate clinical applications with Arzox: In summary, we suggest that the antiestrogenic actions of (a) patients with breast cancer who have never been treated with Arzox on the uterus may translate to clinical practice with the Tam; (b) patients with breast cancer who have been treated with control of occult endometrial cancer growth in patients never Tam previously; and (c) patients who have failed first-line Tam exposed to Tam. In contrast, patients who have been exposed to and have subsequently been switched to a second-line SERM Tam may remain at risk for endometrial cancer during Arzox (Arzox). treatment. We found that in cell culture and in the Tam-naı¨ve ECC-1 transplanted human tumor model, Arzox is effective as an ACKNOWLEDGMENTS antiestrogenic/antitumor agent. Arzox completely antagonized We thank Alexander De Los Reyes for his assistance with the estrogen-stimulated growth in vitro and in vivo. Additionally, animal experiments. Arzox does not completely decrease ER protein levels and is similar to Ral in endometrial cancer cells. ICI182,780 causes REFERENCES destruction of ER (28) and, therefore, has an optimal action with 1. Greenlee, R. T., Murray, T., Bolden, S., and Wingo, P. A. Cancer low intrinsic activity of the ER complex and receptor loss to statistics, 2000. CA Cancer J. Clin., 50: 7–33, 2000. prevent endometrial cancer growth (26). Recently, we have 2. Assikis, V. J., and Jordan, V. C. Gynecologic effects of tamoxifen demonstrated that ICI182,780 given twice a week s.c. (10 mg) and the association with endometrial carcinoma. Int. J. Gynaecol. Ob- has an optimal antitumor effect on the growth of T47D-Tam, stet., 49: 241–257, 1995. MT2-Tam, MCF-7 breast cancers, and EnCa101-Tam endome- 3. Fisher, B., Costantino, J. P., Wickerham, D. L., Redmond, C. K., Kavanah, M., Cronin, W. M., Vogel, V., Robidoux, A., Dimitrov, N., trial cancer (38). The studies of ICI182,780 showing the acti- Atkins, J., Daly, M., Wieand, S., Tan-Chiu, E., Ford, L., and Wolmark, vation of AP-1 in vitro (39) are an artifact that does not translate N. Tamoxifen for prevention of breast cancer: report of the National to therapeutic cross-resistance. The complete degradation of ER Surgical Adjuvant Breast and Bowel Project P-1 Study. J. Natl. Cancer protein is only observed with the pure antiestrogen, whereas Inst., 90: 1371–1388, 1998. SERMs produce different responses. For example, 4-OHT 4. Fisher, B., Costantino, J. P., Redmond, C. K., Fisher, E. R., Wick- erham, D. L., and Cronin, W. M. Endometrial cancer in tamoxifen- slightly increases ER protein in ECC-1 endometrial cells, treated breast cancer patients: findings from the National Surgical Ad- whereas Ral and Arzox decrease ER levels (Fig. 3). In an earlier juvant Breast and Bowel Project (NSABP) B-14. J. Natl. Cancer Inst., study, we demonstrated that the reduction in ER expression with 86: 527–537, 1994. Ral and ICI182,780 does not result from a decrease in ER 5. Phillips, D. H. Understanding the genotoxicity of tamoxifen? Carci- mRNA transcription (22). nogenesis (Lond.)., 22: 839–849, 2001. Neither Tam nor Arzox stimulated ECC-1 tumor growth in 6. Bernstein, L., Deapen, D., Cerhan, J. R., Schwartz, S. M., Liff, J., vivo. This suggests that Arzox would be safe on the uterus in the McGann-Maloney, E., Perlman, J. A., and Ford, L. Tamoxifen therapy for breast cancer and endometrial cancer risk. J. Natl. Cancer Inst., 91: treatment of endocrine therapy-naı¨ve, postmenopausal patients. 1654–1662, 1999. It is not surprising that Tam did not stimulate the growth of 7. Osborne, C. K. Tamoxifen in the treatment of breast cancer. N. Engl. ECC-1 tumors in vivo, because the drug is showing promise as J. Med., 339: 1609–1618, 1998. a hormonal therapy for advanced endometrial cancer (40). Sim- 8. Tamoxifen for early breast cancer: an overview of the randomised ilarly, there is a multicenter, Phase II study showing the efficacy trials. Early Breast Cancer Trialists’ Collaborative Group. Lancet, 351: and safety of Arzox (20 mg p.o./day) in advanced or recurrent 1451–1467, 1998. progestogen-sensitive endometrial cancer (41). 9. Canney, P. A., Griffiths, T., Latief, T. N., and Priestman, T. J. Clinical significance of tamoxifen withdrawal response (Letter). Lancet, Although Arzox is less estrogen-like than Tam in the 1: 36, 1987. rodent uterus (15), we have demonstrated that Arzox is cross- 10. Howell, A., Dodwell, D. J., Anderson, H., and Redford, J. Response resistant with Tam in the EnCa101-Tam tumor model. These after withdrawal of tamoxifen and progestogens in advanced breast data suggest that special care must be taken when monitoring cancer. Ann. Oncol., 3: 611–617, 1992. the uterine effects of Arzox in patients who recently have 11. Takimoto, G. S., Graham, J. D., Jackson, T. A., Tung, L., Powell, received Tam treatment. Additionally, Arzox is equivalent to R. L., Horwitz, L. D., and Horwitz, K. B. Tamoxifen resistant breast Tam in stimulating the growth of an EnCa101 tumor that has cancer: coregulators determine the direction of transcription by antag- onist-occupied steroid receptors. J. Steroid Biochem. Mol. Biol., 69: been exposed previously to Tam (EnCa101-E2 tumors). Clearly, 45–50, 1999. once the Tam-stimulated phenotype has been established, some- 12. Wiebe, V. J., Osborne, C. K., Fuqua, S. A., and DeGregorio, M. W. time in the natural history of a tumor, any SERM, e.g., Ral or Tamoxifen resistance in breast cancer. Crit. Rev. Oncol. Hematol., 14: Arzox, may stimulate growth. On the basis of our present and 173–188, 1993. previous studies with Ral and toremifene (34), (26), we believe 13. Robertson, J. F. Oestrogen receptor: a stable phenotype in breast SERMs cannot inhibit the growth of endometrial cancers that cancer. Br. J. Cancer, 73: 5–12, 1996. are stimulated by Tam, but only the pure antiestrogen 14. Howell, A., DeFriend, D. J., Robertson, J. F., Blamey, R. W., Anderson, L., Anderson, E., Sutcliffe, F. A., and Walton, P. Pharmaco- ICI182,780 (38), which destroys the ER, is able to inhibit such kinetics, pharmacological and anti-tumour effects of the specific anti- tumors. So, in terms of the endometrium, the pure antiestrogen oestrogen ICI 182780 in women with advanced breast cancer. Br. J. seems to be the best option as a second-line hormonal therapy Cancer, 74: 300–308, 1996.

15. Sato, M., Turner, C. H., Wang, T., Adrian, M. D., Rowley, E., and activity in vitro and in vivo. J. Steroid Biochem. Mol. Biol., 59: Bryant, H. U. LY353381.HCl: a novel raloxifene analog with improved 389–396, 1996. SERM potency and efficacy in vivo. J. Pharmacol. Exp. Ther., 287: 1–7, 30. Delmas, P. D., Bjarnason, N. H., Mitlak, B. H., Ravoux, A. C., 1998. Shah, A. S., Huster, W. J., Draper, M., and Christiansen, C. Effects of 16. Baselga, J., Llombart-Cussac, A., Bellet, M., Guillem-porta, V., and raloxifene on bone mineral density, serum cholesterol concentrations, Petruzelka, L. Randomised double-blind phase 2 study of a selective and uterine endometrium in postmenopausal women. N. Engl. J. Med., estrogen receptor modulator (SERM) LY353381 in patients (Pts) with 337: 1641–1647, 1997. locally advanced or metastatic breast cancer (LAMBC). Breast Cancer 31. Cummings, S. R., Eckert, S., Krueger, K. A., Grady, D., Powles, Res Treat., 57: 31, 1999. T. J., Cauley, J. A., Norton, L., Nickelsen, T., Bjarnason, N. H., 17. MacGregor Schafer, J., Lee, E-S., Dardes, R. C., Bentrem, D. J., Morrow, M., Lippman, M. E., Black, D., Glusman, J. E., Costa, A., and O’Regan, R. M., De Los Reyes, A., and Jordan, V. C. Analysis of Jordan, V. C. The effect of raloxifene on risk of breast cancer in cross-resistance of the selective estrogen receptor modulators arzoxifene postmenopausal women: results from the MORE randomized trial. Mul- (LY353381) and LY117018 in tamoxifen-stimulated breast cancer xe- tiple Outcomes of Raloxifene Evaluation. JAMA, 281: 2189–2197, nografts. Clin. Cancer Res., 7: 2505–2512, 2001. 1999. 18. Goldenberg, M., Nezhat, C., Mashiach, S., and Seidman, D. S. A 32. Fornander, T., Rutqvist, L. E., Cedermark, B., Glas, U., Mattsson, randomized, prospective study of endometrial resection to prevent re- A., Silfversward, C., Skoog, L., Somell, A., Theve, T., Wilking, N., et al. Adjuvant tamoxifen in early breast cancer: occurrence of new pri- current endometrial polyps in women with breast cancer receiving mary cancers. Lancet, 1: 117–120, 1989. tamoxifen. J. Am. Assoc. Gynecol. Laparosc., 6: 285–288, 1999. 33. Gottardis, M. M., Robinson, S. P., Satyaswaroop, P. G., and Jordan, 19. Creasman, W. T. Endometrial cancer: incidence, prognostic factors, V. C. Contrasting actions of tamoxifen on endometrial and breast tumor diagnosis, and treatment. Semin. Oncol., 24: S1-140–S1-50, 1997. growth in the athymic mouse. Cancer Res., 48: 812–815, 1988. 20. Horwitz, R. I., Feinstein, A. R., Horwitz, S. M., and Robboy, S. J. 34. O’Regan, R., Gajdos, C., Dardes, R. C., de los Reyes, A., Bentrem, Necropsy diagnosis of endometrial cancer and detection-bias in case/ D., and Jordan, V. C. Effect of raloxifene after tamoxifen on breast and control studies. Lancet, 2: 66–68, 1981. endometrial cancer growth. J. Natl. Cancer Inst., in press, 2001. 21. Satyaswaroop, P. G., Sivarajah, A., Zaino, R. J., and Mortel, R. 35. Buzdar, A. U., Marcus, C., Holmes, F., Hug, V., and Hortobagyi, G. Hormonal control of growth of human endometrial carcinoma in the Phase II evaluation of Ly156758 in metastatic breast cancer. Oncology, nude mouse model. In: F. Bresciane, R. J. B. King, M. Lippman, and 45: 344–345, 1988. J-P. Raynaud (eds.), Progress in Cancer Research and Therapy, vol. 35, 36. Snyder, K. R., Sparano, N., and Malinowski, J. M. Raloxifene pp. 430–435. New York: Raven Press, 1988. hydrochloride. Am. J. Health Syst. Pharm., 57: 1669–1675; quiz 1676– 22. Bentrem, D. J., Dardes, R. C., Liu, H., MacGregor-Schafer, J., Zapf, 1678, 2000. J. W., and Jordan, V. C. Molecular Mechanism of Action at Estrogen ␣ 37. Munster, P. N., Buzdar, A., Dhingra, K., Enas, N., Ni, L., Major, Receptor of a New Clinically Relevant Antiestrogen (GW7604) M., Melemed, A., Seidman, A., Booser, D., Theriault, R., Norton, L., Related to Tamoxifen. Endocrinology, 142: 838–846, 2001. and Hudis, C. Phase I study of a third-generation selective estrogen 23. Pink, J. J., and Jordan, V. C. Models of estrogen receptor regulation receptor modulator, LY353381.HCl, in metastatic breast cancer. J. Clin. by estrogens and antiestrogens in breast cancer cell lines. Cancer Res., Oncol., 19: 2002–2009, 2001. 56: 2321–2330, 1996. 38. Gajdos, C., MacGregor Schafer, J., Dardes, R. C., O’Regan, R. M., 24. Dardes, R. C., O’Regan, R., and Jordan, V. C. Low dose tamoxifen Timm Pearce, S., de los Reyes, A., and Jordan, V. C. Effects of ICI effectively controls endometrial cancer growth not previously exposed 182,780 on tamoxifen-stimulated breast and endometrial cancer. Breast to the drug. Proc. Am. Assoc. Cancer Res., 42: 240, 2001. Cancer Res. Treat., in press, 2001. 25. Clarke, C. L., and Satyaswaroop, P. G. Photoaffinity labeling of the 39. Paech, K., Webb, P., Kuiper, G. G., Nilsson, S., Gustafsson, J., progesterone receptor from human endometrial carcinoma. Cancer Res., Kushner, P. J., and Scanlan, T. S. Differential ligand activation of 45: 5417–5420, 1985. estrogen receptors ER␣ and ER␤ at AP1 sites. Science (Wash. DC), 26. O’Regan, R. M., Cisneros, A., England, G. M., MacGregor, J. I., 277: 1508–1510, 1997. Muenzner, H. D., Assikis, V. J., Bilimoria, M. M., Piette, M., Dragan, 40. Pandya, K. J., Yeap, B. Y., Weiner, L. M., Krook, J. E., Erban, J. K., Y. P., Pitot, H. C., Chatterton, R., and Jordan, V. C. Effects of the Schinella, R. A., and Davis, T. E. Megestrol and tamoxifen in patients antiestrogens tamoxifen, toremifene, and ICI 182,780 on endometrial with advanced endometrial cancer: an Eastern Cooperative Oncology cancer growth. J. Natl. Cancer Inst., 90: 1552–1558, 1998. Group Study (E4882). Am. J. Clin. Oncol., 24: 43–46, 2001. 27. Robinson, S. P., and Jordan, V. C. Antiestrogenic action of 41. Klijn, J. Multicentre Phase II study of the selective estrogen recep- toremifene on hormone-dependent, -independent, and heterogeneous tor modulator (SERM) LY353381 in advanced or recurrent endometrial breast tumor growth in the athymic mouse. Cancer Res., 49: 1758– cancer–objective responses in progestagen sensitive patients. Proc. 1762, 1989. ASCO, 19: 386a, 2000. 28. Parker, M. G. Action of “pure” antiestrogens in inhibiting estrogen 42. Klotz, D. M., Hewitt, S. C., Korach, K. S., and Diaugustine, R. P. receptor action. Breast Cancer Res. Treat., 26: 131–137, 1993. Activation of a uterine insulin-like growth factor I signaling pathway by 29. Somjen, D., Waisman, A., and Kaye, A. M. Tissue selective action clinical and environmental estrogens: requirement of estrogen recep- of tamoxifen methiodide, raloxifene and tamoxifen on creatine kinase B tor-␣. Endocrinology, 141: 3430–3439, 2000.

Downloaded from clincancerres.aacrjournals.org on September 27, 2021. © 2001 American Association for Cancer Research. Effects of the New Selective Estrogen Receptor Modulator LY353381.HCl (Arzoxifene) on Human Endometrial Cancer Growth in Athymic Mice

Rita C. Dardes, David Bentrem, Ruth M. O'Regan, et al.

Clin Cancer Res 2001;7:4149-4155.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/7/12/4149

Cited articles This article cites 38 articles, 8 of which you can access for free at: http://clincancerres.aacrjournals.org/content/7/12/4149.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/7/12/4149.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/7/12/4149. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.