Bazedoxifene Exhibits Antiestrogenic Activity in Animal Models of Tamoxifen-Resistant Breast Cancer: Implications for Treatment of Advanced Disease

Published OnlineFirst March 27, 2013; DOI: 10.1158/1078-0432.CCR-12-3771

Clinical Cancer Cancer Therapy: Preclinical Research

Suzanne E. Wardell, Erik R. Nelson, Christina A. Chao, and Donald P. McDonnell

Abstract Purpose: There is compelling evidence to suggest that drugs that function as pure estrogen receptor (ER-a) antagonists, or that downregulate the expression of ER-a, would have clinical use in the treatment of advanced tamoxifen- and aromatase-resistant breast cancer. Although such compounds are currently in development, we reasoned, based on our understanding of ER-a pharmacology, that there may already exist among the most recently developed selective estrogen receptor modulators (SERM) compounds that would have usage as breast cancer therapeutics. Thus, our objective was to identify among available SERMs those with unique pharmacologic activities and to evaluate their potential clinical use with predictive models of advanced breast cancer. Experimental Design: A validated molecular profiling technology was used to classify clinically relevant SERMs based on their impact on ER-a conformation. The functional consequences of these observed mechanistic differences on (i) gene expression, (ii) receptor stability, and (iii) activity in cellular and animal models of advanced endocrine-resistant breast cancer were assessed. Results: The high-affinity SERM bazedoxifene was shown to function as a pure ER-a antagonist in cellular models of breast cancer and effectively inhibited the growth of both tamoxifen-sensitive and -resistant breast tumor xenografts. Interestingly, bazedoxifene induced a unique conformational change in ER-a that resulted in its proteasomal degradation, although the latter activity was dispensable for its antagonist efficacy. Conclusion: Bazedoxifene was recently approved for use in the European Union for the treatment of osteoporosis and thus may represent a near-term therapeutic option for patients with advanced breast cancer. Clin Cancer Res; 19(9); 1–12. 2013 AACR.

Introduction tamoxifen subsequently respond to aromatase inhibitors The primary goal of endocrine therapies in breast cancer (1–3). This finding highlights the continued dependence a is to block the transcriptional activity of estrogen receptor on ER- signaling within tumors in advanced disease, (ESR1, ER-a) by either (i) inhibiting CYP19A1 (aromatase), raising the possibility that even in tumors that are resistant a the enzyme responsible for the conversion of androgens to to tamoxifen and aromatase inhibitors, the ER- signaling estrogens, or (ii) directly interfering with the transcriptional axis may remain a viable target. activity of the receptor. When used as primary interventions The primary mechanisms underlying resistance to tamox- both approaches are similarly efficacious and improve ifen and aromatase inhibitors are somewhat different. outcome to the same degree. Notable, however, is the However, it now seems that a common feature in either observation that a significant number of patients that case involves the reprograming of ER signaling pathways exhibit de novo or acquired resistance to the antiestrogen within tumor cells, allowing them to continue to capitalize on progrowth and survival pathways downstream of ER-a. The currently available aromatase inhibitors effectively reduce the production of both peripherally and intratumo- Authors' Affiliation: Department of Pharmacology and Cancer Biology, Duke University School of Medicine, Durham, North Carolina rally generated estrogens, and resistance to these agents is not associated with an inability to effectively suppress Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/). estrogen production (4, 5). Rather, there is accumulating evidence that exposure of ER-a–positive breast cancer cells Corresponding Author: Donald P. McDonnell, Duke University Medical Center, Pharmacology and Cancer Biology, Box 3813, Durham, NC 27710. to aromatase inhibitors renders them hypersensitive to Phone: 919-684-6035; Fax: 919-681-7139; E-mail: either residual amounts of steroidal estrogens, dietary/envi- [email protected] ronmental compounds with estrogenic activity, or to doi: 10.1158/1078-0432.CCR-12-3771 endogenously produced molecules that exhibit estrogenic 2013 American Association for Cancer Research. activity but which do not require aromatization (6–8). With

www.aacrjournals.org OF1

Wardell et al.

determined that tamoxifen does not freeze ER-a in an Translational Relevance apo-state but rather it induces a conformational change Breast cancer remains the most commonly diagnosed that enables the presentation of unique protein–protein cancer among women and the leading cause of cancer interaction surfaces on the receptor that dictate its tran- mortality in women worldwide. Although targeted ther- scriptional coregulator-binding preferences (15). Thus, the apies such as tamoxifen and aromatase inhibitors are agonist activity of tamoxifen depends on the relative expres- effective in treating estrogen receptor (ER)–positive sion and/or the activity of functionally distinct coregulators tumors, de novo and acquired resistance remain an in target tissues (16, 17). Although the specific coregulators impediment to durable clinical responses. However, that enable the agonist activity of tamoxifen remain elusive, ER-a remains a therapeutic target in breast cancers that there are considerable additional data to support this are resistant to both first- and second-line endocrine hypothesis. Most notably, by screening for compounds that interventions. Evaluation of the molecular pharmacol- induced a conformational change in ER-a that did not ogy of clinically relevant ER modulators revealed that the present the protein–protein interaction surfaces required third-generation selective estrogen receptor modulator for tamoxifen action, we identified DPC974/GW5638 (SERM) bazedoxifene had a unique mechanism of (18, 19). This compound was shown to be an effective action and inhibited the growth of both tamoxifen- inhibitor of ER action in xenograft models of both tamox- sensitive and -resistant ER-a–positive breast cancer ifen-sensitive and -resistant breast cancers and yielded xenografts. These findings provide strong support for positive results in a heavily treated population of patients the clinical evaluation of bazedoxifene in patients with with breast cancer with endocrine treatment–resistant dis- breast cancer who relapse while undergoing treatment ease (20). Similarly, it has been shown that fulvestrant (ICI with tamoxifen and/or an aromatase inhibitor. 182,780; ICI) also induces a unique conformational change in ER-a, likewise functions as an antagonist in tamoxifen- resistant xenograft models and shows efficacy in patients with advanced disease (15, 21, 22). Together these data respect to the latter, we and others have shown that the suggest that by manipulating ER-a structure and influencing oxysterol 27-hydroxycholesterol, a primary metabolite of coregulator engagement it is possible to develop com- cholesterol, exhibits robust estrogenic activity and is pro- pounds with useful activities in breast cancer. duced at levels that are likely to promote substantial tumor Clearly, there is a very strong rationale to support the growth (ref. 9; unpublished data). Thus, it is anticipated that targeting of ER-a in the setting of tamoxifen- and aroma- high-affinity ER-a antagonists, or compounds that ablate tase-resistant disease. Considering the current state of the ER-a expression, would have clinical activity in aromatase- art in this field, it is likely that drugs with usage in these resistant disease. disease states can be identified by screening for agents that The mechanisms underlying resistance to tamoxifen are (i) bind ER-a with an affinity high enough to outcompete complex. When originally developed, tamoxifen was clas- both endogenous and exogenous estrogens, (ii) exhibit sified as an "antiestrogen," a compound that competitively minimal agonist activity on those genes on which tamox- inhibited the binding of estrogens to ER-a, thus freezing it ifen agonist activity is manifest in resistant breast cancer in an apo-conformation. However, this simple model failed cells, and (iii) induce a structural change in ER-a that to explain how tamoxifen could manifest agonist activity in disables the protein–protein interaction surfaces required bone and in the endometrium and did not explain the for tamoxifen agonist activity. With these criteria in mind, withdrawal responses noted in some patients with breast we conducted a comparative functional analysis of clin- cancer who progressed while on tamoxifen (10–13). It is ically relevant ER ligands, a study that revealed that now known that tamoxifen is not an antiestrogen per se, bazedoxifene, a recently developed high-affinity orally but rather a selective estrogen receptor modulator (SERM), a active SERM, inhibits ER-a action in both tamoxifen- compound whose relative agonist/antagonist activity can sensitive and -resistant xenograft models. This drug was differ between cells. Thus, although it is capable of func- recently approved for use in the European Union for the tioning as an antagonist in the breast, tamoxifen can also treatmentofosteoporosis.The findings presented herein manifest agonist activity in other tissues. This observation should inform near-term clinical studies of this drug in was significant in light of the early studies that suggested patients with advanced breast cancer. that resistance to tamoxifen represented a "switch" within breast cancer cells that enabled them to recognize tamoxifen Materials and Methods as an agonist as opposed to an antagonist (14). Supported Reagents by a wealth of more recent preclinical and clinical data, it ER ligands included 17b-estradiol (E2; Sigma), ICI now seems that it is the ability of breast cancer cells to (Tocris), 4-hydroxytamoxifen (4OHT; Sigma), raloxifene support the agonist activity of tamoxifen that is the primary (Sigma), and endoxifen (Sigma). Lasofoxifene and baze- driver of resistance. The mechanisms underlying the molec- doxifene were gifts from Wyeth, Inc. (now Pfizer). ular pharmacology of tamoxifen in breast cancer are now Ligands were dissolved in ethanol or dimethyl sulfoxide well understood and have been informative with respect to (DMSO). Cycloheximide and MG132 were purchased the development of resistance. Specifically, it has been from Sigma.

OF2 Clin Cancer Res; 19(9) May 1, 2013 Clinical Cancer Research

Bazedoxifene Inhibits ER-Dependent Tumor Growth

Cell culture ovariectomized recipient mice received hormone treatment Cell lines were maintained in Dulbecco’s modified via a timed release pellet (0.72 mg 17b-estradiol or 5 mg Eagle’s medium (DMEM)/F12 (MCF7) or RPMI (BT483, tamoxifen/60 days; Innovative Research of America) rBT474, and SKBR3) media (Invitrogen) supplemented implanted s.c. Two days later, an MCF7 or TamR tumor with 8% FBS (Gemini), nonessential amino acids (Invitro- approximately of 0.8 to 1 cm3 in volume was excised under gen), and sodium pyruvate (Invitrogen). Parental cell lines sterile conditions from a euthanized donor mouse, diced to were obtained from American Type Culture Collection, approximately 2-mm3 sections, and implanted into the which authenticates cell lines by short tandem repeat pro- axial mammary gland under anesthesia (10 g trochar). filing. Lapatinib-resistant rBT474 cells and LTED MCF7 cells Tumor growth was measured 3 times weekly by caliper were maintained as described previously (23, 24). Cells [tumor volume ¼ (A2 B)/2, where A is the longer axis]. were plated for experiments in media lacking phenol red When tumor volume reached approximately 0.2 cm3 mice and supplemented with 8% charcoal-stripped FBS (Gemi- were randomized to continued 17b-estradiol or tamoxifen ni). After 48 hours, cells were treated with ER ligands and/or treatment with placebo or bazedoxifene (5 mg/60 days) inhibitors as indicated and were harvested for immunoblot pellets implanted s.c., or weekly injection with 5 mg/mouse or quantitative real-time PCR (qRT-PCR) analysis 24 hours ICI suspended in corn oil. For the bazedoxifene treatment, after treatment. only group 17b-estradiol pellets were surgically removed simultaneously with bazedoxifene pellet insertion. Immunoblot analysis Protein expression was analyzed as described previously Analysis of tumor tissue (25) using antibodies sc-6259 (cytokeratin 18), sc-20680 Tumors were excised upon reaching approximately 1 cm3 a a (lamin A), sc-5546 ( -tubulin), and sc-8005 (ER- ; Santa and cryopreserved, and frozen tissues were pulverized. Cruz Biotechnology). Protein expression was analyzed essentially as above. RNA was extracted using TRIzol (Invitrogen) as per manufac- RNA isolation and quantitative real-time PCR turer’s instructions, and mRNA expression was detected as RNA isolation and analysis were conducted as described above. previously (26). mRNA abundance was calculated using the DDCT method (27). Primer sequences are available upon request. Results Bazedoxifene induces a unique conformational change Mammalian two-hybrid analysis in ER-a and inhibits its transcriptional activity in Transfections and analysis were conducted as previously cellular models of breast cancer described (26). We and others have shown that tamoxifen partial agonist activity, and by inference tamoxifen resistance, can be Proliferation assays attributed to specific ligand-induced conformational Of note, 3 103 MCF7 cells were plated per well in 96- changes in ER-a that present a unique protein–protein well culture plates. One plate was decanted and frozen on interaction surface(s) that facilitates coregulator recruit- day 1. Remaining plates were treated as indicated on days 1, ment (15). This hypothesis was confirmed in studies show- 4, and 6, with identically treated plates harvested on days 4, ing that compounds that do not present the "tamoxifen 6, and 8. DNA content was detected using a FluoReporter surface" are effective in inhibiting the growth of tamoxifen- assay (Invitrogen) as per manufacturer’s instructions. resistant xenograft models of breast cancer (20). In recent years, several new high-affinity SERMs have been developed Adenovirus production for use in the treatment and prevention of osteoporosis. The Creation, production, and purification of an adenovirus goal of this study was to evaluate whether these clinically expressing human ER-a were previously described (28). relevant SERMs could have activity as breast cancer treat- Forty-eight hours after plating, cells were infected with ments and explore the potential of their near-term use in ER-a adenovirus using multiplicity of infection (MOI) 0 this disease. or 100. Cells were treated with ER ligands as indicated The impact of a series of chemically distinct SERMs on immediately following infection and were harvested for ER-a conformation was evaluated using a previously de- immunoblot or qRT-PCR analysis 24 hours later. For pro- scribed peptide-profiling technology that reads on the dif- liferation assays, cells were infected and treated immediate- ferential availability of protein–protein interactions pre- ly and 2 or 5 days later. sented on the receptor following ligand binding (15, 20). Specifically, peptide probes were selected that could identify Xenograft tumor analyses those surfaces on ER-a that were required for the agonist All procedures were approved by the Duke University activity of estradiol and/or tamoxifen. Using an in-cell two- Institute for Animal Care and Use Committee. 17b-Estra- hybrid assay, we determined that the SERMs bazedoxifene, diol–dependent MCF7 or tamoxifen-stimulated TamR lasofoxifene, and raloxifene facilitate conformational tumors were initiated in the axial mammary gland of female changes in ER-a that are distinct from those observed in NU/NU mice (6 weeks age) by serial transfer. Briefly, the presence of 4OHT or the agonist 17b-estradiol.

www.aacrjournals.org Clin Cancer Res; 19(9) May 1, 2013 OF3

Wardell et al.

A 120,000 Veh B E2 6 100,000 ICI ARHGEF28 5 GW7604 Veh 80,000 4OHT 4 E2 Laso 60,000 Ral 3

activity 40,000 BZA 2 1 20,000 Relative luciferase Relative Relative mRNA level Relative mRNA 0 0 h A o Control SRC ASC2 bN2 β α β α β α bI2 ICI Ral 7 -16 / III / V II Ve BZ Las OHT NRID NRID 4 C 10 4OHT Veh D E 15 F 15 4OHT ICI 8 AGR2 E2 4OHT Laso Ral BZA Veh ICI BZA ICI 6 AGR2 10 10 BZA HSPB8 4 GPR56 2 SNX24 5 5 fluorescence Relative mRNA level mRNARelative RhoBTB1 fluorescence 0 STC1 Fold change in DNA 0 0 Fold change in DNA A ICI SLC9A3R1 veh BZ Ral 10.10.010 0.010.0010 0.1 Laso 4OHT SLC2A1 Treatment (µmol/L) Treatment (µmol/L) KCNK6 INHBB G H I 4OHT 12 veh 5 Veh 10 KRT15 ICI 4OHT 4OHT SERPINA1 10 4 8 BZA BZA ICI RAB37 8 BZA AGR3 ICI 3 6 RASGRP1 6 2 4 MKI67 4 fluorescence

fluorescence fluorescence 2 DDR1 2 1

LRP2 Fold change in DNA Fold change in DNA 0 Fold change in DNA 0 RFTN1 0 86420 0 0.01 10.1 MAOB 86420 Days of treatment Days of treatment Treatment (µmol/L)

Figure 1. Bazedoxifene (BZA) induces a unique conformational change in ER-a and inhibits ER-a activity in cellular models of breast cancer. A, interaction between ER-a and conformation-specific peptides in a mammalian two-hybrid system. Triplicate wells of SKBR3 cells were transfected with plasmids expressing ER-a fused to VP16 together with Gal4DBD alone (control) or Gal4DBD fused to ER-interacting peptides noted on the horizontal axis. Cells were then treated with the indicated ER ligands (100 nmol/L). Interaction of ER-a with the Gal4DBD peptide constructs was detected through activation of a Gal4- responsive luciferase reporter construct and was normalized to detected b-galactosidase activity expressed in a constitutive manner using a second vector. Normalized response is expressed as fold increase over the detected level of interaction between Gal4DBD alone and ER-VP16 in the absence of ligand (Veh). B–D, MCF7 cells were treated for 24 hours with vehicle or 100 nmol/L ligand—ICI, bazedoxifene, raloxifene (Ral), lasofoxifene (Laso), or 4OHT—before RNA isolation and qRT-PCR analysis of the expression of mRNAs shown previously to be responsive to SERMs. mRNA expression was normalized to the expression of the 36B4 housekeeping gene. D, fold changes as compared with vehicle of mRNAs of interest were transformed and normalized as previously described and are presented as a dendrogram constructed in JMP 9.0. E and F, MCF7 cells were plated in phenol red–free media supplemented with charcoal-stripped FBS (E) or in complete media supplemented with FBS (F) and were treated with the indicated ligands (100 nmol/L) on days 1, 4, and 6 of an 8-day proliferation assay. Cells in E were simultaneously treated with 1 nmol/L 17b-estradiol (E2). DNA content as assessed by fluorescence serves as an indicator of cell proliferation. G, MCF7 cells were plated and treated as in E, but were treated with 20 nmol/L insulin instead of 17b-estradiol. H, lapatinib-resistant BT474 cells (rBT474) were plated in complete media supplemented with FBS and with 1 mmol/L GW2974 [EGF receptor (EGFR) inhibitor], and were then treated as in F. I, LTED MCF7 cells were plated in media supplemented with FBS that was stripped of growth factors twice using charcoal. Cells were treated with 0.01 to 1 mmol/L ligands on days 1, 4, and 6 of an 8-day proliferation assay and analyzed as in E. Values (relative increase in DNA fluorescence) in E through I were normalized to values detected in a duplicate plate of cells that were harvested on day 1 before the initial treatment. Data are representative of at least 3 independent experiments.

Furthermore, neither bazedoxifene, lasofoxifene, or ralox- conformation, studies that highlighted the structural ifene occupied ER-a interacted with probes that identified uniqueness of the ER-a–BZA complex (29). surfaces that were uniquely presented on the receptor upon Previously, we conducted a broad survey of the transcrip- binding ICI and GW7604 (Fig. 1A). These results show the tional responses that occur within breast cancer cells when conformational flexibility of ER-a and highlight a potential treated with different SERMs and identified (i) genes regu- opportunity to identify compounds whose impact on recep- lated similarly by all SERMs tested, (ii) genes regulated only tor structure results in a favorable activity in breast cancer. by 4OHT, and (iii) genes whose response differentiated These findings are in agreement with a prior study in which SERMs (26). Using the most informative representative hydrogen/deuterium exchange (HDX) mass spectrometry genes from this study, we evaluated how differences in was used to interrogate the impact of ligands on ER-a ligand-induced changes in ER-a conformation translate

OF4 Clin Cancer Res; 19(9) May 1, 2013 Clinical Cancer Research

Bazedoxifene Inhibits ER-Dependent Tumor Growth

into differences in gene expression. Several genes, were administered bazedoxifene, a regimen intended to ARHGEF28 (RGNEF) for example, respond in a graded reveal any agonist activity of bazedoxifene not evident manner to different SERMs, with 4OHT exhibiting the from the studies conducted in vitro. The results of this greatest agonist activity, raloxifene and bazedoxifene being analysis clearly indicate that both bazedoxifene and ICI significantly less active, and ICI functioning as an inverse significantly inhibit tumor growth rate with a delay in agonist (Fig. 1B). Most notable, however, was the observa- both tumor progression (Fig. 2A and B) and tumor tion that the basal expression and estrogen-dependent doubling time noted (Fig. 2C). Furthermore, when induction of AGR2, a gene associated with breast cancer administered alone, bazedoxifene did not exhibit agonist progression during tamoxifen therapy, is repressed by ICI activity in this model (Fig. 2B). It is important to note that and bazedoxifene but induced by raloxifene, lasofoxifene, ICI was administered at a dose of 5 mg/mouse injected and 4OHT (Fig. 1C). Considering these data, a more com- s.c. once weekly, which translates to a dose approximately prehensive survey of the activity of different SERMs on the 1,000-fold greater than that currently achievable in expression of a large number of genes whose expression was patients with breast cancer. This dose of ICI greatly induced or repressed by tamoxifen was conducted. Analysis exceeds the dose of bazedoxifene used (5 mg/mouse in of the data revealed that among the SERMs studied, the a 60-day continuous release pellet ¼83 mg/kg/d), yet pharmacologic profile of bazedoxifene was most compara- these drugs inhibited tumor growth and reduced tumor ble with the pure antagonist ICI, and thus it was brought doubling time in a similar manner despite being admin- forward for a more complete analysis in relevant models of istered at a dose only 7-fold greater than the daily dose of breast cancer (Fig. 1D). Importantly, it was shown that estradiol. Previously, it has been shown that ICI induces bazedoxifene inhibited estrogen-dependent proliferation turnover of ER-a in breast tumors, and accordingly this of MCF7 cells with efficacy similar to that of ICI and 4OHT drug and others that exhibit similar activities are now (Fig. 1E). As a more stringent test of activity, we assessed classified as selective estrogen receptor degraders (SERD). MCF7 cell proliferation in growth factor–replete FBS media. Given the established SERD activity of ICI, the expression Under these conditions, 4OHT manifests significant partial of ER-a was measured in a subset of tumors harvested at agonist activity, whereas ICI and bazedoxifene were effec- the end of the study. Notwithstanding minor differences tive antagonists (Fig. 1F). It was further shown that 4OHT between tumors in each group, it was noted that the was unable to inhibit insulin-stimulated proliferation of dosing regimen used to deliver ICI affected a substantial, MCF7 cells, whereas both ICI and bazedoxifene effectively although not absolute, turnover of the receptor. Interest- inhibited this activity (Fig. 1G). It was further shown that ingly, we also observed that ER-a expression was similarly ICI and bazedoxifene effectively inhibited the proliferation reduced in bazedoxifene-treated tumors (Fig. 2D). This of lapatinib-resistant rBT474 cells, a tamoxifen-sensitive finding was not totally unexpected as previously we and þ subline of the HER2 BT474 cells in which reactivation of others have shown that bazedoxifene exhibits some SERD ER-a signaling is associated with the development of lapa- activity in vitro (30). tinib resistance (ref. 24; Fig. 1H). Finally, bazedoxifene, 4OHT, and ICI similarly inhibited the proliferation of LTED Bazedoxifene attenuates ER-a–dependent growth of a MCF7 cells (Fig. 1I), a cellular model of resistance to tamoxifen-resistant xenograft tumor model aromatase inhibitors. When taken together, these data sug- The observation that bazedoxifene and ICI functioned gest that the conformational changes induced in ER upon similarly as inhibitors of either tamoxifen- or estradiol- binding ICI or bazedoxifene enable these agents to exhibit induced gene expression in MCF7 cells in vitro led us to favorable activities in relevant models of breast cancer. examine whether bazedoxifene would exhibit efficacy in an These findings encouraged us to conduct a comparative in vivo model of tamoxifen resistance (Fig. 2E). Previously, analysis of these drugs in in vivo models of breast cancer. we reported on the development and characterization of a xenograft model (TamR) in which resistance to tamoxifen is Bazedoxifene attenuates estrogen-dependent growth manifest (20). In brief, in this very stable tumor model, of MCF7-cell–derived tumor xenografts tamoxifen functions as an agonist, whereas both of the As an initial test of the therapeutic potential of baze- SERDs, ICI and DPC974/GW5638, effectively inhibit doxifene, its ability to inhibit the growth of 17b-estradi- growth (20). In contrast with the parental MCF7 tumors, ol–dependent MCF7-derived tumors was compared with which are unable to grow without 17b-estradiol stimulation that of ICI. For these experiments, MCF7-derived tumors (data not shown), TamR tumors grow, albeit slowly, in the were harvested from estrogen-treated donor mice and absence of estradiol supplementation, and their growth is similar-sized tumor fragments were implanted into ovari- dramatically stimulated upon administration of tamoxifen ectomized athymic nu/nu mice. All mice received estro- (5 mg/60-day release pellet ¼83 mg/kg/d; Fig. 3A). Anal- gen supplementation until tumors reached 0.2 cm3 at ysis of tumor volume over time (30 days) indicated that which time the tumor-bearing mice were randomized to bazedoxifene significantly reversed tamoxifen stimulation either of 4 groups. In 3 groups, the estrogen supplemen- of these tumors (Fig. 3B), resulting in a growth rate that was tation was continued (i) alone or together with (ii) equivalent to that observed when tumors were grown in the bazedoxifene or with (iii) ICI. In a fourth group of absence of tamoxifen (Fig. 3C and D). Immunoblot analysis animals, the estradiol pellet was removed and the animals of whole-cell extracts prepared from tumors harvested at the

www.aacrjournals.org Clin Cancer Res; 19(9) May 1, 2013 OF5

Wardell et al.

A B E2 110 ) 3 1.4 100 E2 1.3 E2 + BZA 3 90 E2 + BZA 1.2 E2 + ICI 1.1 80 E2 + ICI 1.0 BZA alone 70 BZA alone 0.9 60 0.8 50 0.7 0.6 40 0.5 30 0.4 20 0.3 10 0.2 % Tumors0.8 cm < 0.1 0 0.0 Average tumor (cm volume 0 10 20 30 40 50 60 70 0 10203040506070 Days to outcome Duration of treatment (d)

C D E2 E2 + BZA E2 + ICI MCF7 WCE E 5 40 AGR2 Vehicle 4 ICI 30 BZA ER 3 20 Rel 6 1 3 4 7 9 8 4 8 9 80 3 33 2 29 2 4 25 2 8 2 Dens 142 1 10

Average tumor 1 doubling time (d) doubling time 0 Lamin A level Relative mRNA 0 E2 +E2 E2 + Rel Veh E2 4OHT BZA ICI 1 3 5 5 7 95 1 97 83 39 14 Dens 1 120 148 1 1 13 101 10

Figure 2. Bazedoxifene attenuates ER and estrogen-dependent growth of MCF7-cell derived tumor xenografts. Mice bearing MCF7 xenograft tumors were randomized (11–12 mice per group) at 0.2 0.025 cm3 tumor volume to receive continued 17b-estradiol (E2) stimulation (17b-estradiol alone), treatment with 17b-estradiol together with either bazedoxifene (s.c. pellet) or ICI (weekly injection) or 17b-estradiol withdrawal (pellet removed) together with bazedoxifene treatment (bazedoxifene alone). A, days required for tumors to reach 0.8 cm3 by Kaplan–Meier analysis. Log-rank test indicated significance (P < 0.0001) in comparison of all treatments to the estrogen control. Differences observed between the E2 þ ICI and E2 þ BZA groups were not significant. B, tumor growth for each group is presented as the average tumor volume for each treatment group SEM at each day of treatment, with the initial day of treatment at randomization considered to be day 0. Nonlinear regression analysis for each growth curve is presented. C, average doubling time for tumors within each group. Data for each replicate were fitted to an exponential growth regression model [Y ¼ Start exp(K X), where we constrained the Start to being shared between all groups]. Doubling time was estimated from these curves and plotted as mean SEM. Nonparametric Kruskal–Wallis analysis followed by the Dunn test indicated significant differences (P < 0.0001) between both groups and 17b-estradiol. D, ER expression detected by immunoblot of whole-cell extracts from tumors harvested from 4 representative mice from each group. Relative density (Rel Dens) of detected bands was calculated by mean densitypixels and normalized to, and expressed as a percentage of, the average density detected for the vehicle-treated samples. E, MCF7 cells were treated for 24 hours with vehicle, 17b-estradiol (1 nmol/L) or 4OHT (10 nmol/L) in the presence or absence of ICI or bazedoxifene (100 nmol/L). AGR2 mRNA expression was detected and normalized as in Fig. 1.

end of the study indicated that ER levels were significantly stimulated growth of these tumors. These findings highlight reduced in those treated with bazedoxifene, whereas, the potential clinical use of bazedoxifene as a breast cancer expectedly, tamoxifen had a minimal effect on receptor therapeutic. expression (Fig. 3E). To further differentiate the actions of bazedoxifene and tamoxifen and to confirm adequate drug Bazedoxifene exhibits SERD activity in cellular models exposure, we assessed the expression of several ER target of ER-a–positive breast cancer genes in tumor mRNA selected for their ability to read on the Bazedoxifene and ICI exhibited similar activities on gene agonist activity of tamoxifen. As expected from our previ- regulation in MCF7 cell-derived tumors and in their TamR ously published work (26), it was determined that the variants, and both compounds resulted in a similar down- mRNAs encoding AGR2, AGR3, and KRT13 were signifi- regulation of ER-a expression in these tumors. Of specific cantly induced in tumors exposed to tamoxifen, and that interest was the observation that treatment of MCF7 cell- this activity was completely inhibited by bazedoxifene. The derived tumors with either bazedoxifene or ICI resulted in a inhibitory activity of tamoxifen on the basal expression of similar downregulation of ER-a expression. The pharma- the estrogen responsive gene C1QTN6 was unaffected by cology of ICI has been probed in detail, and it has been bazedoxifene, indicating, not surprisingly, that these concluded from these studies that this compound induces a drugs are likely to similarly affect some target genes conformational change in the receptor that targets it for (refs. 26, 31, 32; Fig. 3F). These data indicate that bazedox- degradation. Although the conformational change in ER-a ifene does not facilitate tumor growth in a validated model induced by bazedoxifene is distinct from that observed of tamoxifen resistance but rather inhibits tamoxifen- upon binding ICI (Fig. 1A; ref. 29), we were interested in

OF6 Clin Cancer Res; 19(9) May 1, 2013 Clinical Cancer Research

Bazedoxifene Inhibits ER-Dependent Tumor Growth

A 0.8 B Tam 110 Tam 0.7 Tam + BZA None 100 0.6 3 90 0.5 80 )

3 70 0.4 60 (cm 0.3 50 0.2 40 30 0.1

Average tumor volume Average tumor 20

0 % Tumors < 0.5 cm 10 0 20 40 60 0 Days of tumor growth 0 10203040 Days to outcome C D 1.00 Tam ) 3 Tam + BZA 30 0.75 No hormone 25 20 0.50 15 time (d) 0.25 10 5 Average tumor doublingAverage tumor

Average tumor volume (cm 0.00 0 0 10 20 30 40 Tam Tam + No Duration of treatment (d) BZA treatment

E F 2 No hormone Tam Tam + BZA No hormone Tam

MCF7 Tam + BZA 1.5 ER 1 Rel 3 3 Dens 75 97 76 23 2 67 3 30 123 104 2 161 155 177 0.5

α-tubulin level Relative mRNA Rel 0 2 0 61 99 93 88 0 87 8 85 82 Dens 120 101 1 101 126 AGR2 KRT13 C1QTN6 AGR3

Figure 3. Bazedoxifene attenuates ER-dependent growth of tamoxifen-resistant breast tumor xenografts. A, average tumor volume of tamoxifen-resistant xenograft tumors implanted with or without tamoxifen stimulation (5-mg s.c. pellet). B–D, when tamoxifen-stimulated tumors attained 0.2 0.025 cm3 tumor volume, animals were randomized (8–10 mice per group) to receive continued tamoxifen stimulation (tamoxifen alone) or treatment with tamoxifen together with bazedoxifene (5-mg s.c. pellet). B, days required for tumors in each group to reach 0.5 cm3 volume (Kaplan–Meier analysis). Log-rank test indicated significant difference (P ¼ 0.0175). C, tumor growth for each group is presented as average tumor volume SEM per study arm at each day of treatment, with the initial day of treatment at randomization considered to be day 0. Tumor volume for the unstimulated (no hormone) group is plotted corresponding to the first day of treatment for the bazedoxifene groups. Nonlinear regression analysis for each growth curve is presented. D, average doubling time for tumors in each group. Data for each replicate were fitted to an exponential growth regression model [Y ¼ Start exp(K X), where we constrained the Start to being shared between all groups]. Doubling time was estimated from these curves and plotted as mean SEM. Nonparametric Mann–Whitney test indicated a significant difference (P ¼ 0.0207) between tamoxifen and Tam þ BZA. E, ER expression detected by immunoblot of whole-cell extracts of tumors harvested from 4 representative mice from each group. Relative density (Rel Dens) was calculated as in Fig. 2. F, average expression level ( SEM) of tamoxifen-regulated genes in tumors (n ¼ 8–10) from each treatment group. mRNA levels were detected and normalized to human 36B4 as in Fig. 1. Data are plotted relative to the tamoxifen-treated group. determining if the downregulation of ER-a observed in receptor. For comparative purposes, and to validate the cell bazedoxifene-treated tumors reflects SERD activity, as is the models, we showed that treatment of cells with 17b-estra- case for ICI, or occurs in an indirect manner (i.e., decreased diol resulted in a quantitative downregulation of ER-a ER-a mRNA expression). Notably, the impact of ligands on expression. ER-a stability/expression observed in tumors was recapit- We next explored the mechanisms underlying the appar- ulated in cellular models of breast cancer. As shown, treat- ent SERD activity of bazedoxifene. First, it was shown that ment of MCF7 (Fig. 4A) or BT483 (Fig. 4B) cells for 8 or 24 the effects of bazedoxifene, ICI, or 17b-estradiol on ER-a hours with bazedoxifene or ICI resulted in a loss of ER-a expression occurred in a posttranscriptional manner as no expression, whereas 4OHT, endoxifen, tamoxifen, raloxi- differences in ER-a mRNA expression were observed fol- fene, and lasofoxifene had little effect on the level of the lowing drug treatment (Fig. 4C). It was further shown that

www.aacrjournals.org Clin Cancer Res; 19(9) May 1, 2013 OF7

Wardell et al.

A B 4OHT Tam Endox Ral Vehicle E2 ICI BZA Laso E2 BZA Vehicle ICI Ral Laso

8 h 8 h

Rel Rel 6 Dens 14 66 26 Dens 31 10 68 43 100 153 141 123 139 100 101

24 h 24 h Rel Rel 5 0 3 0 5 3 3 Dens 19 71 6 3 75 Dens 1 65 28 100 127 11 100 11

C D 1.5 E2 ICI BZA Ral 4OHT ER-α Vehicle Vehicle 1.0 2 h Rel 0 2 2 9 4 h Dens 13 1 3 9 0.5 10 104 1

expression 24 h

Relative mRNARelative CHX 0.0 Rel 0 2 8 4 10.110.10.1 0 1 39 86 5 Dens 1 1 ICIE2Veh BZA MG132 [µmol/L treatment] Rel 8 6 8 3 Dens 2 8 79 100 15 1 13 1

Figure 4. Bazedoxifene reduces ER-a expression in ER-a–positive breast cancer cells. MCF7 (A) and BT483 (B) breast cancer cells were treated for 8 or 24 hours with 17b-estradiol (100 nmol/L) or ER antagonists (1 mmol/L). ER-a levels in these whole-cell extracts were analyzed by immunoblot. C, MCF7 cells were pretreated 2 hours with MG132 (30 mm) or cycloheximide (CHX; 10 mg/mL) before 4 hours treatment with ER ligands (100 nmol/L 17b-estradiol, 1 mmol/L others). ER expression was analyzed by Western immunoblot analysis of whole-cell extracts. D, MCF7 cells were treated for 2, 4, or 24 hours with 0.1 or 1 mmol/L 17b-estradiol, bazedoxifene, or ICI before RNA isolation and detection of ER-a mRNA by qRT-PCR. ER mRNA levels were normalized to the housekeeping gene 36B4. Loading controls for panels A and B are shown in Supplementary Fig. S1. Relative density (Rel Dens) of blot images was calculated as in Fig. 2, but normalized to, and expressed as a percentage of, the density detected for the vehicle-treated control.

ligand-dependent downregulation of ER-a expression was saturating, maximally efficacious doses of each compound, not affected by cotreatment with cycloheximide, suggesting it was noticed that, unlike what occurs in ICI or GW7604 that receptor stability was an intrinsic property of the treated cells, ER-a expression was not completely down- receptor–ligand complex and did not require the de novo regulated in cells treated with bazedoxifene (Fig. 5B). This synthesis of ancillary proteins required for receptor turn- raised the interesting question as to the requirement of over. Furthermore, 17b-estradiol-, ICI-, and bazedoxifene- receptor turnover for maximal antagonist activity, an mediated turnover of the receptor could be blocked important issue to address as it speaks to the use of ER-a by cotreatment of cells with the proteasome inhibitor measurements in tumors as a surrogate for the antiestro- MG132 (Fig. 4D). Together, these results suggest that, genic actions of bazedoxifene. similar to ICI, bazedoxifene exhibits proteasome-depen- Because proteasomal degradation of ER-a, when occu- dent SERD activity. pied by different ligands, requires it to engage a specific E3 ligase(s), it stands to reason that if the expression of these Inhibition of ER-a turnover does not compromise the latter enzymes is limiting that it should be possible to antagonist activity of bazedoxifene saturate the turnover process by overexpression of the Having shown that bazedoxifene binding induces ER-a receptor. Previously, we showed that the ubiquitination turnover, we next conducted a comparative analysis of this patterns of 17b-estradiol- and ICI-occupied ER-a in cells compound with ICI and GW7604, another SERD that has were different (33, 34), suggesting that they may be targeted been evaluated previously in patients with breast cancer. by different E3 ligases. In follow-up experiments, it was When evaluated for their ability to inhibit 17b-estradiol– shown that ICI-mediated turnover of ER-a was reduced mediated induction of the transcription of GREB1 mRNA in upon receptor overexpression (35). Under the same con- MCF7 cells, it was observed that all 3 compounds were ditions, however, a quantitative downregulation of 17b- equally effective as antagonists with potencies that reflected estradiol–occupied receptor was observed. Thus, the ability their affinity for the receptor (Fig. 5A); similar results were to saturate the degradation process by overexpressing ER-a observed on other ER-responsive genes (26). However, at afforded us the opportunity to assess whether turnover of

OF8 Clin Cancer Res; 19(9) May 1, 2013 Clinical Cancer Research

Bazedoxifene Inhibits ER-Dependent Tumor Growth

A B C GREB1 ICI 50 BZA (Veh) Veh GW7604 Veh 0 MOI = E2 ICI BZA Ral 4OHT MCF7 E2 GW7604 ICI Veh GW7604 BZA 40 MOI 30 ER ns 0 20 Rel Rel 9 9 9 8 Dens 23 28 Dens 14 37 72 10 100 100 115 100 cyto 18 100 Relative mRNA level Relative mRNA 0 00.010.11 Rel Rel 0 6 2 2 5 0 1 0 Dens 0 77 0 3 Dens 36 9 6 7 1 0 SERM (µmol/L) 1 1 118 1 211 1 2 264 343 2 3 1

D 5 E 6 TFF1 MOI 0 Veh IGFBP4 MOI 0 Veh MOI 0 E2 4 MOI 0 E2 5 MOI 100 Veh MOI 100 Veh 4 3 MOI 100 E2 MOI 100 E2 3 2 level level 2 1 1 Relative mRNA Relative mRNA 0 0 LasoRalBZAICIVeh 4OHT Veh ICI LasoRalBZA 4OHT

F G J Veh E2 25 Veh 25 Veh BZA BZA 20 ICI 20 ICI Veh E2 BZA ICI 15 4OHT 15 4OHT MOI 4OHT Day 10 10 0 2 fluorescence 5 fluorescence 5 Rel 0 8 Dens 62 36 0 46 10 1 Fold change in DNAFold change in 0 DNAFold change in 0 7520 7520 MOI 0 MOI 0 Days of treatment Days of treatment 100 2 Rel 1 H I Dens 00 06 Veh E2 1 263 100 3 25 Veh 25 Veh BZA 20 BZA 20 ICI ICI 100 5 15 4OHT 15 4OHT Rel 0 5 Dens 0 79 99 10 10 1 157 fluorescence 5 fluorescence 5 100 7 Fold change in DNA 0 Fold change in DNA 0 Rel 9 5 7520 7520 Dens 79 9 100 157 MOI 100 Days of treatment MOI 100 Days of treatment

Figure 5. Bazedoxifene-dependent inhibition of ER-mediated transcriptional activity is not affected by loss of ER turnover. A, MCF7 cells were treated for 24 hours with or without 17b-estradiol (E2; 1 nmol/L) in the presence of the indicated SERDs (0.01–1 mmol/L). mRNA expression of target gene GREB1 was detected as in Fig. 1. B, MCF7 cells were treated for 24 hours with 17b-estradiol (100 nmol/L) or SERDs (1 mmol/L), and ER-a levels were analyzed by immunoblot. C–E, MCF7 cells were infected with ER-a adenovirus (MOI 100) or mock infected before 24 hours treatment with 17b-estradiol (100 nmol/L) or other ligands (1 mmol/L). C, ER-a levels were analyzed by immunoblotting of whole-cell extracts. D and E, expression of ER target genes in cells infected in parallel and treated 24 hours with 17b-estradiol (10 nmol/L) in the presence or absence of SERMs or ICI (1 mmol/L) was analyzed by qRT-PCR as in Fig. 1. F–I, MCF7 cells were infected as in C, with 0 (F and G) or 100 (H and I) MOI of adenovirus expressing ER-a and treated with ligands (1 mmol/L) in the presence or absence of 17b-estradiol (1 nmol/L) immediately, and 2 and 5 days following infection, with one replicate plate harvested before infection and 2, 5, and 7 days after infection. Cell proliferation in the absence (F and H) or presence (G and I) of estrogen was analyzed as in Fig. 1. J, MCF7 cells were infected, treated with ligands, and harvested in parallel with F through I. ER-a expression was analyzed as in C. Data are representative of at least 3 independent experiments. Relative density (Rel Dens) of blot images was calculated as in Fig. 2, but normalized to, and expressed as a percentage of, the density detected for the vehicle treated control. For C, values are normalized to, and expressed as a percentage of, the vehicle-treated uninfected sample. In H, values are normalized to, and expressed as a percentage of, the vehicle control for each MOI and time.

www.aacrjournals.org Clin Cancer Res; 19(9) May 1, 2013 OF9

Wardell et al.

the receptor was required for the antagonist activity of the reproductive tract, in that its use was associated with bazedoxifene. As shown in Fig. 5C, the anticipated effects significantly increased endometrial thickness and increased of the various SERMs/SERDs on ER-a stability were incidence of endometrial polyps, uterine leiomyoma, and observed in uninfected cells (MOI ¼ 0). However, in cells vaginal bleeding (44). Lasofoxifene has been approved in infected with a virus expressing ER-a, up to 10-fold over- the European Union for the treatment and prevention of expression of the receptor was achieved, and under these osteoporosis but has not yet been registered. Recently, conditions the SERD activity of ICI, bazedoxifene, and bazedoxifene was approved and marketed in the European GW7604 was saturated. We have previously shown that Union for the same use. Our studies of the molecular receptor overexpressed in this manner is active and able to pharmacology of bazedoxifene and lasofoxifene indicate bind ligand (35). As observed previously, 17b-estradiol– that the former most closely profiled with fulvestrant (ICI), mediated turnover of ER-a was not affected by any of these an approved SERD. Importantly, as shown in this study, manipulations, ruling out nonspecific effects of this viral bazedoxifene effectively inhibited the growth of tamoxifen- overexpression protocol on the activity of the proteasome. sensitive and -resistant breast cancer xenografts. Given these The most important finding, however, is that overexpres- data, and the fact that this drug is currently available for use sion of ER-a had no significant effects on the antagonist in humans, we believe that a near-term evaluation of its efficacy of the SERMs/SERDs evaluated in this study. As efficacy in patients with advanced disease is justified. expected, overexpression of ER-a did lead to an increase in One of the unexpected findings made in this study is that ligand-independent activation of transcription by the recep- bazedoxifene induces ER-a turnover both in vitro and in vivo tor; however, this activity was attenuated by all of the SERMs absent any impact on the expression of ER-a mRNA. The and SERDs tested (Fig. 5D and E). Furthermore, the inhi- results observed in vivo are of particular note as they indicate bition of ER-a function by SERMs and SERDs, as measured that a sustained knockdown of ER-a protein expression can by comparing the proliferation of infected and uninfected be accomplished by bazedoxifene over time and suggest the cells, was unaffected as well despite sustained overexpres- absence of feedback mechanism(s) to restore receptor sion of ER-a (Fig. 5F–J). When taken together, these results levels. When considering the mechanisms by which resis- suggest that although the removal of ER-a by an ER ligand is tance to endocrine therapies can arise, it is clear that it would a desirable attribute of antagonists, it is not required for the be advantageous to chronically suppress ER-a expression. antagonist activity of the existing SERDs. In this manner, ER-a, the target for coactivators and signaling pathways that impinge upon the ER-coregulator com- Discussion plex(s), is removed and signaling is attenuated. Important- In recent years, there has been a resurgence of interest in ly, we have shown that although bazedoxifene manifests ER-a as a therapeutic in cancer. In part, this has been fueled SERD activity, receptor degradation is not required for its by recent advances that have been made in the pharmaceu- antagonist activity. This is an important finding as receptor tical exploitation of the androgen receptor (AR) in prostate overexpression has been considered as a possible explana- cancer. Until relatively recently, it was considered that the tion for resistance to tamoxifen in breast cancer. AR was not a viable target in castrate-resistant prostate Our studies have revealed that ICI and bazedoxifene cancer. However, the spectacular clinical responses to exhibit very similar pharmacology and function similarly Cyp17 inhibitors, such as abiraterone, and to the new in relevant animal models of breast cancer, findings that high-affinity AR antagonist MDV3100 in late-stage disease, provide a rationale for its clinical evaluation. However, have encouraged a reappraisal of ER as a therapeutic target there has been significant concern of late as to the use in those breast cancers that have failed both tamoxifen and of ICI in breast cancer. Although this drug is approved for aromatase inhibitors (36, 37). As with AR, our understand- the treatment of tamoxifen/aromatase-inhibitor–refractory ing of the molecular pharmacology of ER has advanced tumors, its response rate as originally determined in the tremendously in recent years, enabling the development EFECT (Evaluation of Faslodex vs. Exemestane) trial was of highly predictive mechanism-based screens for ER mod- approximately 10%, similar to intervention with exemes- ulators. Indeed, lasofoxifene and bazedoxifene emerged tane as a second line, steroidal aromatase inhibitor (21). from in vitro screens that selected for compounds that This low response rate was unexpected and was initially (i) bound ER-a and competitively displaced 17b-estradiol, considered to indicate that SERD intervention would have (ii) inhibited estrogen action in cellular models of breast limited use in the treatment of advanced breast cancer. cancer, and (iii) did not manifest ER-a agonist activity in However, a considerable amount of additional data has contexts where tamoxifen functioned as an agonist (38–40). emerged of late indicating that the pharmacokinetic prop- In clinical trials, it was determined that both of these erties of fulvestrant, administered by intramuscular injec- compounds exhibited robust estrogenic activity in bone tion, and not its mechanism of action, is likely that which and reduced the incidence of vertebral fractures by approx- limits its efficacy (45, 46). Indeed, despite the use of a imately 40% in osteoporotic postmenopausal patients (41, loading dose 3 times higher than that evaluated in the 42). In addition, as a secondary endpoint, lasofoxifene was EFECT trial, drug concentrations measured in the vicinity shown to significantly decrease the risk of ER-positive breast of the tumor were found to be insufficient to saturate the cancer (43). However, lasofoxifene was subsequently found receptor (47). A compelling series of positron emission to have a less favorable clinical profile than bazedoxifene in tomography imaging studies confirmed poor delivery of

OF10 Clin Cancer Res; 19(9) May 1, 2013 Clinical Cancer Research

Bazedoxifene Inhibits ER-Dependent Tumor Growth

fulvestrant to the tumor (43). Regardless, the results of the Disclosure of Potential Conflicts of Interest CONFIRM (comparison of fulvestrant in recurrent or met- S.E. Wardell is a consultant/advisory board member of Pfizer and Novo- lipid, Inc. E.R. Nelson is a consultant/advisory board member of NovoLipid, astatic breast cancer) trial suggested that this SERD may Inc. D.P. McDonnell has a commercial research grant from Pfizer and is a have improved biologic activity and clinical efficacy at consultant/advisory board member of the same. No potential conflicts of higher doses than that originally approved (48). Specifical- interest were disclosed by the other author. ly, it was noted that in the high-dose arm of this study, Authors' Contributions fulvestrant significantly delayed time to cancer progression Conception and design: S.E. Wardell, E.R. Nelson, D.P. McDonnell compared with standard-dose regimens. However, despite Development of methodology: S.E. Wardell, E.R. Nelson, D.P. McDonnell these advances in the delivery of fulvestrant, it is likely that Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): S.E. Wardell, E.R. Nelson, C.A. Chao the full therapeutic potential of SERD intervention has not Analysis and interpretation of data (e.g., statistical analysis, biosta- been realized with this drug. Being a high-affinity, orally tistics, computational analysis): S.E. Wardell, E.R. Nelson, D.P. bioavailable drug, bazedoxifene would not be limited by McDonnell Writing, review, and/or revision of the manuscript: S.E. Wardell, E.R. poor tumor access. Thus, its bioavailability presents the Nelson, C.A. Chao, D.P. McDonnell opportunity to evaluate the true potential of SERD activity Administrative, technical, or material support (i.e., reporting or orga- nizing data, constructing databases): S.E. Wardell, E.R. Nelson, C.A. Chao in late-stage disease. Study supervision: S.E. Wardell, D.P. McDonnell Our studies introduce an apparent paradox in that bazedoxifene, originally developed as a SERM for use in Acknowledgments the treatment and prevention of osteoporosis, actually Lapatinib-resistant rBT474 cells and long-term estrogen-deprived (LTED) cells were kindly provided by Drs. Neil Spector (Duke University, Durham, profiles as a pure antagonist/SERD in the breast. The latter NC) and Richard Santen (University of Virginia), respectively. implies that in the context of bone, bazedoxifene exhibits ER-a agonist activity in bone, a finding that at first Grant Support glance is incompatible with it being a SERD. Indeed, This study was financially supported by Pfizer Pharmaceuticals, Inc. and DPC974/GW5638, an earlier molecule we developed that DK048807. The costs of publication of this article were defrayed in part by the exhibited substantial SERD activity, is also bone protec- payment of page charges. This article must therefore be hereby marked tive (19). This raises the possibility that within the appro- advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate priate cells in bone ER-a stability is not affected in the this fact. same manner as in breast cancer cells, a hypothesis that Received December 10, 2012; revised February 18, 2013; accepted March we are currently testing. 19, 2013; published OnlineFirst March 27, 2013.

References 1. Kvinnsland S, Anker G, Dirix LY, Bonneterre J, Prove AM, Wilking N, 10. Dallenbach-Hellweg G, Schmidt D, Hellberg P, Bourne T, Kreuzwieser et al. High activity and tolerability demonstrated for exemestane in E, Doren M, et al. The endometrium in breast cancer patients on postmenopausal women with metastatic breast cancer who had tamoxifen. Arch Gynecol Obstet 2000;263:170–7. previously failed on tamoxifen treatment. Eur J Cancer 2000;36: 11. Farnell YZ, Ing NH. Endometrial effects of selective estrogen receptor 976–82. modulators (SERMs) on estradiol-responsive gene expression are 2. Dodwell D, Wardley A, Johnston S. Postmenopausal advanced breast gene and cell-specific. J Steroid Biochem Mol Biol 2003;84:513–26. cancer: options for therapy after tamoxifen and aromatase inhibitors. 12. Love RR, Mazess RB, Barden HS, Epstein S, Newcomb PA, Jordan VC, Breast 2006;15:584–94. et al. Effects of tamoxifen on bone mineral density in postmenopausal 3. Goss PE, Ingle JN, Martino S, Robert NJ, Muss HB, Piccart MJ, et al. A women with breast cancer. N Engl J Med 1992;326:852–6. randomized trial of letrozole in postmenopausal women after five years 13. Canney PA, Griffiths T, Latief TN, Priestman TJ. Clinical significance of of tamoxifen therapy for early-stage breast cancer. N Engl J Med 2003; tamoxifen withdrawal response. Lancet 1987;1:36. 349:1793–802. 14. Gottardis MM, Jordan VC. Development of tamoxifen-stimulated 4. Miller WR, Larionov AA. Understanding the mechanisms of aromatase growth of MCF-7 tumors in athymic mice after long-term antiestrogen inhibitor resistance. Breast Cancer Res 2012;14:201. administration. Cancer Res 1988;48:5183–7. 5. Brodie A, Jelovac D, Sabnis G, Long B, Macedo L, Goloubeva O. Model 15. Norris J, Paige L, Christensen D, Chang C, Huacani M, Fan D, et al. systems: mechanisms involved in the loss of sensitivity to letrozole. J Peptide antagonists of the human estrogen receptor. Science 1999; Steroid Biochem Mol Biol 2005;95:41–8. 285:744–6. 6. Masri S, Phung S, Wang X, Wu X, Yuan YC, Wagman L, et al. Genome- 16. Kressler D, Hock MB, Kralli A. Coactivators PGC-1 beta and SRC-1 wide analysis of aromatase inhibitor-resistant, tamoxifen-resistant, interact functionally to promote the agonist activity of the selective and long-term estrogen-deprived cells reveals a role for estrogen estrogen receptor modulator tamoxifen. J Biol Chem 2007;282: receptor. Cancer Res 2008;68:4910–8. 26897–907. 7. Jelovac D, Macedo L, Goloubeva OG, Handratta V, Brodie AM. 17. Smith CL, Nawaz Z, O'Malley BW. Coactivator and corepressor Additive antitumor effect of aromatase inhibitor letrozole and anties- regulation of the agonist/antagonist activity of the mixed antiestrogen, trogen fulvestrant in a postmenopausal breast cancer model. Cancer 4-hydroxytamoxifen. Mol Endocrinol 1997;11:657–66. Res 2005;65:5439–44. 18. Bentrem DJ, Dardes RC, Liu H, Maccgregor-Schafer J, Zapf JW, 8. Santen R, Jeng M-H, Wang J-P, Song R, Masamura S, McPherson R, Jordan VC. Molecular mechanism of action at estrogen receptor alpha et al. Adaptive hypersensitivity to estradiol: potential mechanism for of a new clinically relevant antiestrogen (GW7604) related to tamoxifen. secondary hormonal responses in breast cancer patients. J Steroid Endocrinology 2001;142:838–46. Biochem Mol Biol 2001;79:115–25. 19. Willson TM, Norris JD, Wagner BL, Asplin I, Baer P, Brown HR, et al. 9. DuSell CD, Umetani M, Shaul PW, Mangelsdorf DJ, McDonnell DP. 27- Dissection of the molecular mechanism of action of GW5638, a novel Hydroxycholesterol is an endogenous selective estrogen receptor estrogen receptor ligand, provides insights into the role of ER in bone. modulator. Mol Endocrinol 2008;22:65–77. Endocrinology 1997;138:3901–11.

www.aacrjournals.org Clin Cancer Res; 19(9) May 1, 2013 OF11

Wardell et al.

20. Connor CE, Norris JD, Broadwater G, Willson TM, Gottardis MM, 34. Wijayaratne A, Nagel S, Paige L, Christensen D, Norris J, Fowlkes D, Dewhirst MW, et al. Circumventing tamoxifen resistance in et al. Comparative analysis of mechanistic difference among anties- breast cancers using antiestrogens that induce unique confor- trogens. Endocrinology 1999;140:5828–40. mational changes in the estrogen receptor. Cancer Res 2001;61: 35. Wardell SE, Marks J, McDonnell DM. The turnover of estrogen receptor 2917–22. a by the selective estrogen receptor degrader (SERD) fulvestrant is a 21. Chia S, Gradishar W, Mauriac L, Bines J, Amant F, Federico M, et al. saturable process that is not required for antagonist efficacy. Biochem Double-blind, randomized placebo controlled trial of fulvestrant com- Pharmacol 2011;82:122–30. pared with exemestane after prior nonsteroidal aromatase inhibitor 36. Scher HI, Fizazi K, Saad F, Taplin ME, Sternberg CN, Miller K, et al. therapy in postmenopausal women with hormone receptor-positive, Increased survival with enzalutamide in prostate cancer after chemo- advanced breast cancer: results from EFECT. J Clin Oncol 2008;26: therapy. N Engl J Med 2012;367:1187–97. 1664–70. 37. de Bono JS, Logothetis CJ, Molina A, Fizazi K, North S, Chu L, et al. 22. Gottardis MM, Jiang SY, Jeng MH, Jordan VC. Inhibition of tamoxifen- Abiraterone and increased survival in metastatic prostate cancer. N stimulated growth of an MCF-7 tumor variant in athymic mice by novel Engl J Med 2011;364:1995–2005. steroidal antiestrogens. Cancer Res 1989;49:4090–3. 38. Ke HZ, Qi H, Crawford DT, Chidsey-Frink KL, Simmons HA, Thompson 23. Masamura S, Santner SJ, Heitjan DF, Santen RJ. Estrogen deprivation DD. Lasofoxifene (CP-336,156), a selective estrogen receptor mod- causes estradiol hypersensitivity in human breast cancer cells. J Clin ulator, prevents bone loss induced by aging and orchidectomy in the Endocrinol Metab 1995;80:2918–25. adult rat. Endocrinology 2000;141:1338–44. 24. Xia W, Bacus S, Hegde P, Husain I, Strum J, Liu L, et al. A model of 39. Komm BS, Kharode YP, Bodine PV, Harris HA, Miller CP, Lyttle CR. acquired autoresistance to a potent ErbB2 tyrosine kinase inhibitor and Bazedoxifene acetate: a selective estrogen receptor modulator with a therapeutic strategy to prevent its onset in breast cancer. Proc Natl improved selectivity. Endocrinology 2005;146:3999–4008. Acad Sci U S A 2006;103:7795–800. 40. Crabtree JS, Peano BJ, Zhang X, Komm BS, Winneker RC, Harris HA. 25. Wittmann B, Sherk A, McDonnell D. Definition of functionally important Activity of three selective estrogen receptor modulators on hormone- mechanistic differences among selective estrogen receptor down- dependent responses in the mouse uterus and mammary gland. Mol regulators. Cancer Res 2007;67:9549–60. Cell Endocrinol 2008;287:40–6. 26. Wardell SE, Kazmin D, McDonnell DP. Research resource: transcrip- 41. Silverman SL, Christiansen C, Genant HK, Vukicevic S, Zanchetta JR, tional profiling in a cellular model of breast cancer reveals functional de Villiers TJ, et al. Efficacy of bazedoxifene in reducing new vertebral and mechanistic differences between clinically relevant SERM and fracture risk in postmenopausal women with osteoporosis: results between SERM/estrogen complexes. Mol Endocrinol 2012;26: from a 3-year, randomized, placebo-, and active-controlled clinical 1235–48. trial. J Bone Miner Res 2008;23:1923–34. 27. Livak K, Schmittgen T. Analysis of relative gene expression data using 42. Cummings SR, Ensrud K, Delmas PD, LaCroix AZ, Vukicevic S, Reid real-time quantitative PCR and the 2(DDC(T)) method. Methods DM, et al. Lasofoxifene in postmenopausal women with osteoporosis. 2001;25:402–8. N Engl J Med 2010;362:686–96. 28. Gaillard S, Grasfeder LL, Haeffele CL, Lobenhofer EK, Chu TM, 43. LaCroix AZ, Powles T, Osborne CK, Wolter K, Thompson JR, Thomp- Wolfinger R, et al. Receptor-selective coactivators as tools to define son DD, et al. Breast cancer incidence in the randomized PEARL trial of the biology of specific receptor-coactivator pairs. Mol Cell 2006;24: lasofoxifene in postmenopausal osteoporotic women. J Natl Cancer 797–803. Inst 2010;102:1706–15. 29. Dai SY, Chalmers MJ, Bruning J, Bramlett KS, Osborne HE, Montrose- 44. Goldstein SR, Neven P, Cummings S, Colgan T, Runowicz CD, Krpan Rafizadeh C, et al. Prediction of the tissue-specificity of selective D, et al. Postmenopausal evaluation and risk reduction with lasofox- estrogen receptor modulators by using a single biochemical method. ifene (PEARL) trial: 5-year gynecological outcomes. Menopause Proc Natl Acad Sci U S A 2008;105:7171–6. 2011;18:17–22. 30. Lewis-Wambi JS, Kim H, Curpan R, Grigg R, Sarker MA, Jordan VC. 45. Linden HM, Kurland BF, Peterson LM, Schubert EK, Gralow JR, Specht The selective estrogen receptor modulator bazedoxifene inhibits hor- JM, et al. Fluoroestradiol positron emission tomography reveals dif- mone-independent breast cancer cell growth and down-regulates ferences in pharmacodynamics of aromatase inhibitors, tamoxifen, estrogen receptor alpha and cyclin D1. Mol Pharmacol 2011;80: and fulvestrant in patients with metastatic breast cancer. Clin Cancer 610–20. Res 2011;17:4799–805. 31. Creighton CJ, Massarweh S, Huang S, Tsimelzon A, Hilsenbeck SG, 46. Wakeling A, Dukes M, Bowler J. A potent specific pure antiestrogen Osborne CK, et al. Development of resistance to targeted therapies with clinical potential. Cancer Res 1991;51:3867–73. transforms the clinically associated molecular profile subtype of breast 47. Robertson J, Nicholson R, Bundred N, Anderson E, Rayter Z, Dowsett tumor xenografts. Cancer Res 2008;68:7493–501. M, et al. Comparison of the short-term biological effects of 7alpha-[9- 32. Massarweh S, Osborne CK, Creighton CJ, Qin L, Tsimelzon A, Huang (4,4,5,5,5-pentafluoropentylsulfinyl)-nonyl]estra-1,3,5, (10)-triene- S, et al. Tamoxifen resistance in breast tumors is driven by growth 3,17beta-diol (Faslodex) versus tamoxifen in postmenopausal women factor receptor signaling with repression of classic estrogen receptor with primary breast cancer. Cancer Res 2001;61:6739–46. genomic function. Cancer Res 2008;68:826–33. 48. Di Leo A, Jerusalem G, Petruzelka L, Torres R, Bondarenko I, Khasanov 33. Wijayaratne A, McDonnell D. The human estrogen receptor-a is a R, et al. Results of the CONFIRM phase III trial comparing fulvestrant ubiquitinated protein whose stability is affected differentially by ago- 250 mg with fulvestrant 500 mg in postmenopausal women with nists, antagonists, and selective estrogen receptor modulators. J Biol estrogen receptor positive advanced breast cancer. J Clin Oncol Chem 2001;276:35684–92. 2010;28:4594–600.

OF12 Clin Cancer Res; 19(9) May 1, 2013 Clinical Cancer Research

Bazedoxifene Exhibits Antiestrogenic Activity in Animal Models of Tamoxifen-Resistant Breast Cancer: Implications for Treatment of Advanced Disease

Suzanne E. Wardell, Erik R. Nelson, Christina A. Chao, et al.

Clin Cancer Res Published OnlineFirst March 27, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-12-3771

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2013/03/28/1078-0432.CCR-12-3771.DC1

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/early/2013/04/17/1078-0432.CCR-12-3771. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.