Rational management of endocrine resistance in breast cancer

A comprehensive review of estrogen receptor biology, treatment options, and future directions


  • Sara A. Hurvitz MD,

    Corresponding author
    1. Department of Medicine, Division of Hematology-Oncology, University of California-Los Angeles School of Medicine, Los Angeles, California
    • Department of Medicine, Division of Hematology-Oncology, UCLA School of Medicine, Box 957077, 10945 Le Conte Avenue, PVUB Suite 3360, Los Angeles, CA 90095-7077===

    Search for more papers by this author
    • Fax: (310) 829-6192

  • Richard J. Pietras PhD, MD

    1. Department of Medicine, Division of Hematology-Oncology, University of California-Los Angeles School of Medicine, Los Angeles, California
    Search for more papers by this author


Endocrine therapy for breast cancer was introduced more than 100 years ago. In the last 30 years, it has been demonstrated that tamoxifen significantly improves outcomes for patients with hormone-responsive breast tumors. Aromatase inhibitors, which suppress the production of estrogen, are recognized today as an effective alternative for estrogen-receptor-positive breast cancer in postmenopausal women. However, despite an initial response to treatment, many tumors eventually recur or progress. When selecting subsequent endocrine therapy, it is helpful to understand the mechanisms of hormone resistance, consider the goals of treatment, and evaluate the clinical potential of each available drug. The objective of this article was to review the underlying mechanisms of action and resistance for each type of hormone therapy, evaluate the most recent data regarding the use of endocrine agents after disease progression or recurrence, and explore potential combinations of hormone therapies with novel molecules that target key growth factor signaling pathways. Cancer 2008. © 2008 American Cancer Society.

The influence of ovarian hormones on breast cancer was appreciated first more than 100 years ago, when Beatson demonstrated that inoperable breast tumors regress after oophorectomy.1 Since then, pharmacologic antiestrogen measures have been developed and have been shown to significantly improve clinical outcomes for patients with both early- and late-stage hormone receptor-positive (HR+) breast cancer. Because approximately 75% of postmenopausal women with breast cancer have tumors that are positive for estrogen receptor (ER) expression, antiestrogen therapy plays a role in disease management for the majority of patients with breast cancer.2 Today, 2 classes of antihormone endocrine agents are considered the standard of care for primary treatment of early- or late-stage HR+ breast cancer: tamoxifen, which blocks ER activity within tumor cells, and aromatase inhibitors (AIs), such as anastrozole, letrozole, or exemestane, which inhibit the production of estrogen through the aromatase enzyme pathway. Whereas tamoxifen is effective in both premenopausal and postmenopausal women, AIs are indicated only in women who are postmenopausal. In the adjuvant setting, several studies have demonstrated that AIs are tolerated well and offer improved disease-free survival compared with tamoxifen3–5 or placebo.6 The benefits of AIs also have been demonstrated in the metastatic7–10 and neoadjuvant settings.11–14

Despite an initial response to hormone therapy (whether in the adjuvant or metastatic setting), many patients will progress during therapy. Although it is standard practice in recurrent or progressive HR+ breast cancer to switch to a different endocrine agent, many questions remain unanswered regarding the appropriate choice of a subsequent agent.15, 16 In any event, the goals of treatment for patients with metastatic breast cancer are to provide clinical benefit (response, stable disease) with the aim of prolonging life, delaying disease progression, maintaining quality of life (QOL), and postponing the use of cytotoxic chemotherapy that may further impair QOL. The objectives of this article are to review the underlying mechanisms of action and resistance for each type of hormone therapy, evaluate recent data regarding the use of subsequent endocrine agents after recurrence or progression on initial treatment, and explore the future directions of hormone agents.

Mechanisms of Estrogen Receptor Activation

The classic mechanism of steroid hormone action involves estrogen binding to ERs in the nucleus, thereby promoting association with specific estrogen-response elements in the promoter region of target genes.17 However, it is well documented that significant populations of ERs may be located outside the nucleus18 and that ERs also regulate the expression of many genes without directly binding to DNA but by interacting with other transcription factors, such as activator protein 1, and with extranuclear signaling complexes that modulate downstream gene expression.19

When they are not bound by estrogen, ERs often interact with coregulator proteins that modulate ER activity20 and/or with chaperone proteins that stabilize the receptor or hide its DNA-binding domain. There are 2 major subtypes of ERs—ERα and ERβ—each with different isoforms and splice variants.21 ERα is expressed predominantly as a 66-kD transcript in breast tumors,22 whereas ERβ occurs as the product of a different gene.18 ERβ has structural homology to ERα, particularly in the DNA-binding domain (95% amino acid identity) and in the ligand-binding domain (55% amino acid identity).23 Ongoing work indicates that ERβ can modulate ERα activity, suggesting that ERβ and its splice variants may affect the responsiveness of breast cancer to estradiol.24

ERs are activated by 2 general mechanisms: ligand-dependent activation (the classic pathway), in which estrogen binds to the ER and the resulting estrogen-ER unit then interacts directly with DNA to regulate gene transcription; and ligand-independent activation, in which an ER is activated, in part, after phosphorylation by growth factor receptors or other molecules with serine or tyrosine kinase domains (Fig. 1).18 Both ligand-dependent and ligiand-independent ER activation can lead to the modulation of downstream intracellular signaling cascades.18, 21 In this way, activation of the ER pathway ultimately may lead to tumor progression and the proliferation of tumor-associated vascular endothelial cells.25, 26

Figure 1.

Summary of dependent and independent signaling pathways in estrogen receptor (ER)-positive tumor cells.18 ARO indicates aromatase; E2, estradiol; EGF, epidermal growth factor; ARG, amphiregulin; TGFα, transforming growth factor alpha; HRG, heregulin; eNOS, endothelial nitric oxide synthase; AKT, protein kinase C; PI3K, phosphatidylinositol 3-kinase; Src, sarcoma oncogene; VEGF, vascular endothelial growth factor; EGFR, epidermal growth factor receptor; HER2, human epidermal growth factor receptor 2; IGFR1, insulin-like growth factor receptor 1; Shc, Src homology 2 domain-containing transforming protein; Grb2, growth factor receptor bound protein 2; SOS, son of sevenless; MEK, mitogen-activated protein kinase/extracellular signal-regulated kinase; MNAR, modulator of nongenomic actions of the estrogen receptor; G, G protein; MAPK, mitogen-activated protein kinase; IKK, IκB kinase; mTOR, mammalian target of rapamycin; PLC, phospholipase C, cAMP, cyclic adenosine monophosphate; P, protein, Co Reg, coregulator proteins; ERE, estrogen-response element. Reprinted with permission granted by the American Association for Cancer Research and the authors R. J. Pietras and D. C. Marquez-Garban from Pietras RJ, Marquez-Garban DC. Membrane-associated estrogen receptor signaling pathways in human cancers. Clin Cancer Res. 2007;13(16):4672-4676.

Mechanisms of Resistance to Hormone Therapy

With increased understanding of the complex, interconnecting signaling pathways that regulate cellular responses to estrogen has come a realization that tumor cells may take a multitude of different avenues to become resistant to antiestrogen therapy. Unraveling this complexity is essential for optimizing the use and sequencing of currently available drugs and for developing new and more potent ones.

Resistance to therapy may be described either as intrinsic or de novo (a tumor does not respond to a drug from the onset of therapy) or as acquired (a tumor that initially responded to therapy resumes growing).27 One potential cause of intrinsic resistance to tamoxifen is a genetic polymorphism in the enzyme pathway involved in metabolizing tamoxifen to its more active metabolite, endoxifen. Results from pharmacogenetic studies have revealed that patients treated with tamoxifen who have certain polymorphisms in the cytochrome P450 2D6 (CYP2D6) gene (leading to a lower level of endoxifen) may have a higher risk of recurrence.28, 29 Although those results are intriguing, additional studies are needed to assess possible relations between this genotype and resistance to tamoxifen therapy. Whereas genetic polymorphisms reflect innate differences in tamoxifen responsiveness, there also are many potential mechanisms of acquired resistance, some theoretical and others supported by preclinical or clinical data (Table 1).27, 30, 31 Antiestrogen-resistant tumors may develop through clonal selection of tumor cell subsets with altered HR expression, response to tamoxifen, ER structure, or activation of ER pathways.

Table 1. Potential Mechanisms of Acquired Resistance to Endocrine Therapy*
ER/PR Status at RelapseMechanism of ResistanceAffected AgentsClinical Relevance
  • ER indicates estrogen receptor; PR, progesterone receptor; TAM, tamoxifen; +, positive; SERMs, selective estrogen-receptor modulators; AIs, aromatase inhibitors; −, negative; EGFR, epidermal growth factor receptor; HER-2/neu, human epidermal growth factor receptor 2.

  • *

    Reprinted with permission by Lippincott Williams & Wilkins Wolters Kluwer Health for this use only from Johnston SR. Acquired tamoxifen resistance in human breast cancer—potential mechanisms and clinical implications. Anticancer Drugs. 1997;8(10):911-930.27

ER+Metabolic tolerance (by drug exclusion or sequestration) in hormone-sensitive cellsTAMTheoretically, tumor could respond to increased TAM doses; little supporting evidence
ER+ER+ cell stimulation by agonist component of TAM or its metabolitesTAMNovel antiestrogens (eg, newer SERMs) may remain active
ER+Constitutively active or inactive ER mutants or variantsTAMLittle clinical evidence that ER mutants play a major role in resistance
ER+Tumor remodeling with ER+ cell clones that have altered sensitivity/response to TAMTAMMay remain sensitive to other endocrine therapies (eg, AIs)
ER+Activation of ER-regulated growth pathways independent of steroid control (also called adaptive estrogen hypersensitivity)TAM, AIsTumor likely to be completely endocrine resistant; may be caused by long-term AI use
ER−/PR+Clonal selection of ER− cells from an original ER+ tumorTAMMore likely in metastatic recurrences after adjuvant TAM
ER− or ER low+/PR+Overexpression of growth factors or their receptors (eg, EGFR, HER-2/neu, Akt)TAMLikely that these tumors develop complete resistance to endocrine therapy; associated with ligand-independent signaling
ER−/PR−Clonal selection of PR− cells from an original ER+ tumorTAM, AIsData from clinical trials suggest that TAM and AIs are similarly ineffective in these patients

Altered hormone receptor expression

Whereas loss of ERα expression has been demonstrated in 17% to 28% of patients who have acquired resistance to tamoxifen, mutations in the ERα receptor rarely are found, suggesting that this is an unlikely mechanism of tamoxifen resistance.30 The finding that ER expression is maintained in the majority of tamoxifen-resistant tumors suggests that the unresponsive phenotype is caused by a complex, multifactorial change in the expression of a network of genes rather than a simple, single-gene effect. This also may explain why approximately 66% of patients who develop recurrent disease on tamoxifen will respond to AIs or to fulvestrant.31 Loss of progesterone receptor (PR) expression, possibly related to sustained stimulation of growth factor-signaling pathways that suppress expression of the PR gene, appears to be another mechanism leading to tamoxifen resistance, but not to AI resistance.30, 32 Preclinical and clinical data also indicate that increased expression of ER coregulator molecules may increase the agonistic effect of tamoxifen, contributing to resistance.30 Evidence exists that estrogen-independent growth may be associated with a high level of ERα,33 and others have reported changes in expression levels of ERα but little change in ERβ expression in tamoxifen- and fulvestrant-resistant tumors.34

Altered crosstalk between estrogen receptors and growth factors

Altered expression of growth factors and signaling proteins also may contribute to the antiestrogen-resistant phenotype. Candidate molecules include epidermal growth factor (EGF) receptor (EGFR), human epidermal growth factor receptor 2 (HER-2/neu), tumor necrosis factor alpha, protein kinase C alpha (PKCα), mitogen-activated protein kinase (MAPK) phosphatase 3, and p21-activated kinase 1.35–39 Evidence suggests that receptor tyrosine kinase signaling has an impact on resistance to endocrine therapy by modulating the subcellular localization of ER coregulators.40 Preclinical data suggest that ERα function is maintained in the tamoxifen-resistant MCF-7 breast cancer cell line through EGFR/MAPK-mediated signaling.41

Data from preclinical studies and from retrospective analyses of clinical trials indicate that HER-2/neu overexpression is a negative predictor of response to tamoxifen and, to a lesser extent, AIs.42–44 Changes in the insulin-like growth factor 1 receptor (IGF-1R)/IGF-1 pathway also may be responsible for resistance to antiestrogen therapy.42 Understanding the precise mechanisms and outcomes of crosstalk between ERs and these molecules remains incomplete; however, ligand-independent activation of ER may play an important role. The complexity and redundancy observed in these interconnected pathways suggest that concurrent or sequential inhibition of multiple pathways may be a better strategy to improve response to endocrine therapy and reduce the development of resistance.18, 45, 46

Effects of long-term estrogen deprivation

Human breast cancer cells that are deprived of estradiol adapt by developing estrogen hypersensitivity (Fig. 2).47, 48 This may explain the clinical observation in which hormone-dependent breast cancer that initially regressed after oophorectomy-induced estradiol deprivation in premenopausal women regrew in response to low-dose estradiol and regressed further after exposure to AIs. Additional studies in breast cancer cell lines subjected to long-term estrogen-deprivation (LTED) demonstrated that these cells first became hypersensitive to low-dose estrogen and then became estrogen independent.49 The acquired hypersensitivity appeared to be linked to increased expression of ERα and ERβ and was accompanied by a ligand-independent increase in ERα phosphorylation.49 Results from several in vitro and in vivo studies suggest that tumors undergo 3 stages that lead to tamoxifen resistance: In stage 1, tamoxifen acts as an estrogen antagonist; in stage 2, the tumor increasingly becomes sensitive to the agonistic (proestrogenic) effects of tamoxifen; and, in stage 3, the tumor has increased sensitivity (hypersensitivity) to estradiol (as shown in Fig. 3).50, 51 Long-term exposure of breast cancer to tamoxifen, thus, may lead to estrogen hypersensitivity, which, in turn, creates an environment that selects for the growth of tumors with tamoxifen resistance.52 The possibility that LTED with tamoxifen leads to estrogen hypersensitivity bolsters the rationale for using agents that do not have proestrogenic activity (eg, AIs, fulvestrant) for breast cancer that has progressed or recurred on tamoxifen.52 Therefore, the concept of adaptive hypersensitivity to estradiol and the biologic mechanisms underlying this phenomenon have important clinical implications.

Figure 2.

Dose-response effects of estradiol on wild-type (WT) (solid line) and long-term estrogen-deprived (LTED) (hatched line) MCF-7 cells. The leftward shift in the response to estradiol in the LTED cell line demonstrates the concept of estrogen hypersensitivity.47 Reprinted with permission granted by the Endocrine Society to reproduce this figure from Masamura S, Santner SJ, Heitjan DF, Santen RJ. Estrogen deprivation causes estradiol hypersensitivity in human breast cancer cells. J Clin Endocrinol Metab. 1995;80:2918-2925.

Figure 3.

Effects of estradiol (E2) on MCF-7 human breast cancer tumor xenograft (A) and uterine (B) weight in mice after long-term (5-month) treatment with tamoxifen (black bars) or vehicle (gray bars).51 Reprinted from Berstein LM, Wang JP, Zheng H, Yue W, Conaway M, Santen RJ. Long-term exposure to tamoxifen induces hypersensitivity to estradiol. Clin Cancer Res. 2004;10(4):1530-1534. Reprinted with permission granted by the American Association for Cancer Research and the authors.

Biologic Basis of Sequential Endocrine Therapy

After resistance to initial endocrine therapy for breast cancer develops, no clear treatment guidelines exist regarding subsequent hormone treatment for patients with recurrent or systemic disease. Although National Comprehensive Cancer Network guidelines recommend second-line hormone therapy (nonsteroidal or steroidal AIs; fulvestrant or other selective estrogen-receptor modulators [SERMs]; estrogen antagonists; and estrogens, progestins, or androgens) in patients with systemic disease, the guidelines do not recommend a preferred treatment or the appropriate sequencing of agents in this setting.15

The remaining critical issues faced by clinicians and their patients are the timing and choice of second-line therapy. A primary consideration is the nature of the response to initial hormone therapy, because it is an excellent predictor of subsequent response, even when a patient's tumor becomes refractory to current hormone therapy. An estimated 40% to 50% of patients with breast cancer who had a response to initial hormone therapy will respond to subsequent treatment with other hormone agents.53

Other important considerations include the efficacy and safety profiles of available options and issues related to patient compliance with treatment. Because of possible cross-resistance, it is also important to select a drug for subsequent treatment that has a different mechanism of action from the initial therapy.53 Minimally toxic endocrine therapies usually are preferred over cytotoxic chemotherapy given its greater potential for serious adverse events (AEs) and effects on QOL.15

Treatment options after initial therapy with an aromatase inhibitor and/or tamoxifen

The optimal sequence of hormone therapy remains to be established in randomized, controlled clinical trials. However, depending on the initial therapy, we describe current options after treatment failure. Whereas tamoxifen historically has been the first-choice therapy for postmenopausal women with ER+ early breast cancer; now, it is surpassed increasingly by the AIs administered either alone or in sequence with tamoxifen, because AIs produce a significant benefit in the primary adjuvant setting.54 In patients who were treated previously with tamoxifen, an AI can be used as subsequent therapy. Several trials demonstrated that AIs significantly prolong survival55, 56 and time to disease progression55–57 versus megestrol acetate when used as second-line therapy in advanced ER+ breast cancer after failure on first-line tamoxifen. However, the converse (tamoxifen as second-line therapy after an AI) may not be viable in some patients because of the potential for tamoxifen resistance in LTED cells. Although some evidence suggests that the AIs may differ in terms of their pharmacokinetics58 and potencies,59 the biologic meaning and clinical utility of these findings remain unresolved. It is noteworthy that there is some evidence in favor of fulvestrant as a viable option in this situation, because it induces antitumor effects in LTED cells.60

Fulvestrant as an option after failure on aromatase inhibitors or tamoxifen

Antagonist versus Agonist Properties

Like tamoxifen, fulvestrant completely inhibits binding of estradiol to the ER, although fulvestrant has a higher affinity than tamoxifen for the ER.61 Unlike tamoxifen, fulvestrant induces a conformational change in the ER and inhibits receptor dimerization, rendering the complex transcriptionally inactive. Fulvestrant binding also reduces nuclear uptake of the drug-receptor complex, inhibits coactivator recruitment and ER binding to estrogen-responsive genes, and enhances rapid degradation of the receptor. Thus, fulvestrant is considered a pure estrogen antagonist with no agonist activity. Unlike tamoxifen or the AIs, fulvestrant completely inhibits ER signaling.61 In addition, fulvestrant, as an ER down-regulator, has a unique biologic mechanism of action compared with other available hormone agents, including AIs and tamoxifen. There are several pathways involved in ER activation, as detailed above. The finding that fulvestrant directly targets the ER and leads to both cessation of ER signaling and ER degradation may prevent or postpone resistant disease.62–64

Studies in MCF-7 breast cancer cells in vitro and in vivo have demonstrated that, although LTED cells are refractory to treatment with tamoxifen, they are sensitive to treatment with fulvestrant (Fig. 4).49, 60 These cells reportedly had enhanced ERα and IGF-IR/insulin receptor substrate-2 expression and signaling, supporting a role for these molecules in the stimulation of tumor growth by tamoxifen and suggesting a mechanism for their susceptibility to fulvestrant.60 These experiments provided the rationale for the clinical development of fulvestrant as a treatment for HR+ breast cancer that is refractory to antiestrogen therapy.

Figure 4.

Dose-response curves for fulvestrant (A) and tamoxifen (B) in wild-type and long-term estrogen-deprived (LTED) high (LTED-H) and intermediate (LTED-I) cells.49, 60 (A) Reprinted with permission by the Society for Endocrinology as it appeared in Martin LA, Pancholi S, Chan CM, Farmer I, Kimberley C, Dowsett M, Johnston SR. The anti-oestrogen ICI 182,780, but not tamoxifen, inhibits the growth of MCF-7 breast cancer cells refractory to long-term oestrogen deprivation through down-regulation of oestrogen receptor and IGF signaling. Endocr Relat Cancer. 2005;12(4):1017-1036. Copyright Society for Endocrinology 2005. Reproduced by permission. (B) Reprinted from the Journal of Steroid Biochemistry and Molecular Biology with permission from Elsevier Ltd. as it appeared in Chan CM, Martin LA, Johnston SR, Ali S, Dowsett M. Molecular changes associated with the acquisition of oestrogen hypersensitivity in MCF-7 breast cancer cells on long-term oestrogen deprivation. J Steroid Biochem Mol Biol. 2002;81(4-5):333-341. Copyright 2002 Elsevier.

Further evidence of the lack of agonistic effects of fulvestrant comes from clinical trials results in healthy volunteers and women with breast cancer. Results from a phase 1 trial in healthy postmenopausal volunteers demonstrated that, compared with placebo, fulvestrant administered as a single intramuscular injection did not have estrogenic effects on the endometrium and significantly inhibited estrogen-stimulated thickening of the endometrium.65 When administered as a monthly intramuscular injection in women with advanced breast cancer, fulvestrant had minor effects on serum hormones and lipid levels and did not induce or aggravate hot flushes, sweats, or vaginal dryness, all of which are AEs frequently associated with tamoxifen treatments.66

Efficacy and safety of fulvestrant

Overall, phase 3 trials have demonstrated comparable efficacy and tolerability between fulvestrant and anastrozole (after tamoxifen failure) or exemestane (after nonsteroidal AI failure).

Phase 3 comparison of fulvestrant versus anastrozole

The efficacy of fulvestrant versus anastrozole was established in 2 randomized, controlled clinical trials (1 conducted in North America and 1 conducted in Europe, Australia, and South Africa) in postmenopausal women with locally advanced or metastatic breast cancer.67–69 In a prospectively planned, combined analysis of those trials, a total of 851 postmenopausal women who had recurrent or progressive disease on endocrine therapy (>95% had received prior tamoxifen) were randomized to receive fulvestrant or anastrozole until disease progression or early study withdrawal.67, 69 After a median follow-up of 15.1 months, the median time to progression (TTP) was 5.5 months for fulvestrant and 4.1 months for anastrozole (P = .48).67 The overall response rates (ORRs) in patients with visceral metastases were 18.8% for fulvestrant (complete response [CR] rate, 10.1%) and 14% for anastrozole (CR rate, 1.2%; P = .43).69 The median duration of response in patients who responded to fulvestrant (CRs and partial responses) was 16.7 months versus 13.7 months for those who responded to anastrozole. In addition, a retrospective analysis indicated that 43 of 82 patients (52%) who responded to fulvestrant maintained their response for >12 months versus 30 of 70 patients (43%) for anastrozole (P = .16).70 With an extended median follow-up of 27 months, the median overall survival (OS) was 27.4 months for the fulvestrant treatment arm and 27.7 months for the anastrozole group (hazard ratio, 0.98; P = .809).68

Fulvestrant and anastrozole were tolerated similarly well. The only AE category that differed between the 2 groups was joint disorders, which were more frequent in the anastrozole arm (P = .0036).67

Phase 3 comparison of fulvestrant versus exemestane

The Evaluation of Faslodex versus Exemestane Clinical Trial (EFECT), a randomized, double-blind, multicenter, phase 3 trial, demonstrated similar benefits with fulvestrant loading dose (500 mg on Day 0, followed by 250 mg on Days 14 and 28, with 250 mg monthly thereafter) versus exemestane 25 mg daily in postmenopausal women with HR-positive advanced breast cancer that progressed or recurred after nonsteroidal AI therapy.71

In the EFECT trial, the median TTP was 3.7 months for both treatment groups (hazard ratio, 0.963; 95% confidence interval, 0.819-1.133; P = .6531).71 The ORR (7.4% for fulvestrant vs 6.7% for exemestane; P = .736) and the clinical benefit rate (CBR), which was defined as the ORR and stable disease that lasted ≥24 weeks, also were similar between the 2 groups (32.2% for fulvestrant vs 31.5% for exemestane; P = .853).71 The median response duration from the time of randomization was 13.5 months for fulvestrant and 9.8 months for exemestane.71 More recently, at a median follow-up of 20.9 months, OS did not differ significantly between the fulvestrant and exemestane treatment arms (P = .9072; median OS, 24.3 months vs 23.1 months, respectively).71 The 2 treatment arms had similar AE profiles, with only 2% and 2.6% of fulvestrant- and exemestane-treated patients withdrawing because of an AE, respectively.71 These data support the efficacy and safety of fulvestrant (loading dose) and exemestane as hormone therapy after AI failure.

In addition to the approved dose, alternate dosing strategies, including the loading dose and high dose, were used in several recently completed trials (eg, EFECT; Neoadjuvant Endocrine Therapy for Women With Estrogen-Sensitive Tumors [NEWEST]) and currently are under investigation in ongoing fulvestrant trials (eg, the COmparison of Faslodex In Recurrent or Metastatic breast cancer [CONFIRM] trial and the FINDER clinical fulvestrant trials) to further elucidate the optimal fulvestrant dosing strategy.72 Pharmacokinetic evidence suggests that using the loading dose substantially shortens the time to steady-state levels of fulvestrant (shortened to approximately 1 month vs 3 to 6 months with the approved dose).71, 73 Despite this, it remains unclear whether use of the loading dose confers additional clinical benefits. However, recent data from the NEWEST study were the first to demonstrate superior efficacy of high-dose fulvestrant (500-mg loading dose followed by 500 mg per month) compared with the approved dose (250 mg per month).74 That study reported significant reductions in the Ki67 cell proliferation marker (P < .0001) and in ER expression (P < .0003) in patients who received high-dose fulvestrant compared with patients who received the approved dose (P < .0001).75 In addition, the overall tumor response rate at Week 4 of treatment was 17.4% in the high-dose group and 11.8% in the low-dose group.76

Use of steroidal aromatase inhibitors after progression on nonsteroidal aromatase inhibitors

A single-arm, phase 2 study demonstrated that exemestane had activity in patients with metastatic breast cancer after failure of prior nonsteroidal AIs.77 Patients in that study received exemestane 25 mg per day orally; at the time of disease progression, they were offered dose escalation to exemestane 100 mg per day. Among the 242 enrolled patients, the ORR was 6.6%, 17.4% of patients had stable disease that lasted ≥24 weeks, and there was an overall CBR of 24.3% for third-line or fourth-line exemestane.77 These data suggest, perhaps surprisingly, that exemestane has incomplete cross-resistance with nonsteroidal AIs. The efficacy of exemestane after failure of a nonsteroidal AI also was supported in a retrospective study of 114 patients with metastatic breast cancer.78 In that analysis, the CBR of exemestane was 46% (33% in patients with visceral disease), and the median progression-free survival and OS were 18 weeks and 61 weeks, respectively.

Low-dose and high-dose estradiol

Results of in vivo studies have demonstrated the intriguing antitumor potential of low-dose estradiol in tamoxifen-stimulated tumors. Investigators led by V. Craig Jordan developed an ovariectomized, athymic nude mouse model with serial xenotransplants to study exposure of breast cancer cells to 5 years of tamoxifen adjuvant therapy. By using this model, Jordan and collaborators observed that breast tumors appear to cycle through different stages of hormone dependency during the course of long-term tamoxifen exposure.79

During 1 of these stages, after stopping tamoxifen treatment, the administration of physiologic doses of estradiol was tumoricidal, leading to a more effective reduction in tumor size than that observed with stopping tamoxifen alone. Later, some of the breast cancer cells reverted to an estrogen-dependent phenotype that, once again, was sensitive to tamoxifen inhibition. The authors proposed a cyclic model of breast cancer hormone dependency and postulated that it may be possible to maintain patients on tamoxifen therapy with intermittent estrogen treatment.79

Further preclinical investigations by the same laboratory led to the proposal of a cyclic clinical therapeutic strategy of alternating treatment with tamoxifen, AIs, or fulvestrant with brief periods of low-dose estrogen to prevent the development of resistance and, thus, forestall the use of cytotoxic chemotherapy.80, 81 Those authors identified 2 phases of resistance: the first phase, in which the tumor has become resistant to tamoxifen but is susceptible to AIs or fulvestrant; and, after long-term tamoxifen treatment, the second phase, in which the tumor growth is stimulated by tamoxifen but inhibited by estrogen.82 Clinical development of this strategy is intriguing and would be served best by identification and validation of biologic markers to distinguish between first- and second-phase resistance. Limited trials of estrogen treatment for advanced, resistant breast malignancies currently are ongoing in the clinic.83, 84

It is noteworthy that updated analyses of an early randomized trial of diethylstilbestrol (DES) (a potent synthetic estrogen) versus tamoxifen in postmenopausal women with metastatic breast cancer revealed that long-term survival was significantly better for women who received DES than for women who received tamoxifen (35% for the DES arm vs 16% for the tamoxifen arm).85 The basis of this survival advantage is not known. Although tamoxifen remains a preferred agent in the treatment of metastatic breast cancer, that trial highlighted the finding that high doses of estrogenic agents may have significant activity and may play a role in the treatment of selected patients with metastatic breast cancer.

Progestins and androgens

Early studies demonstrated the comparable efficacy of the progestin megestrol acetate and tamoxifen in women with metastatic breast cancer.86 Subsequent studies have focused on optimizing the dosing of this agent in advanced breast cancer. Megestrol acetate still is used by many oncologists in this setting, because it has demonstrated antitumor efficacy; and, as a side effect, it stimulates appetite and causes weight gain.87 Because of this history, additional progestin-based therapies are being evaluated. The 2 pathways for estrogen formation in breast tumors are the aromatase pathway, which is targeted by AIs, and the sulfatase pathway, which converts estrogen sulfate into estrone.88 Interest in new clinical applications of progestins may resurface because of experimental evidence indicating high activity of sulfatase in breast cancer tumors (especially from hormone-dependent cell lines) and the possibility of inhibiting this pathway with progestins. Results of preclinical studies have indicated that certain progestins are strong inhibitors of sulfatase activity in hormone-dependent breast cancer cells. It also has been noted that progestins inhibit 17β-hydroxysteroid dehydrogenase conversion of estrone to estradiol and stimulate sulfotransferase synthesis of estrogen sulfate.88 Data from studies in humans are very limited. Perhaps future clinical trials will determine whether progestins, possibly in association with AIs, may have a role in the treatment of hormone-sensitive breast cancer. Additional studies suggest that progestins also may stimulate tumor progression by promoting the expression of vascular endothelial growth factor (VEGF), which, in turn, acts to facilitate tumor-associated angiogenesis, underscoring the need for more research in this area.89

Although some previous reports suggested a role for androgen therapy in reducing breast cancer progression after failure on antiestrogen therapy, more recent clinical trials have not demonstrated superiority of tamoxifen plus fluoxymesterone over tamoxifen alone as adjuvant therapy for postmenopausal women with resected, early breast cancer with known ER+ status.90 Androgens (ie, testosterone, fluoxymesterone, testolactone, and calusterone) currently are used in some patients with advanced ER+ cancer and have been associated with responses and symptom relief in approximately 20% of patients. However, androgens are associated with a high rate of adverse effects, including masculinization, hair loss, increased libido, and edema.91

Future Directions

Combinations of endocrine therapies currently are under investigation in an effort to improve treatment strategies for postmenopausal patients with advanced or metastatic breast cancer (Table 2).72 The Faslodex and Arimidex in Combination Trial and Southwest Oncology Group Study S0226 are multicenter, phase 3 studies comparing single-agent anastrozole with anastrozole plus fulvestrant in postmenopausal women with ER+/PR+ breast cancer that recurs after the primary treatment of localized tumor. The Study of Faslodex, Exemestane, and Arimidex Trial is comparing fulvestrant monotherapy versus fulvestrant plus anastrozole versus exemestane monotherapy in patients with advanced or metastatic breast cancer that progressed on a nonsteroidal AI.

Table 2. Ongoing Trials of Combinations of Hormone Therapies in Postmenopausal Women With Estrogen Receptor-Positive, Advanced Breast Cancer*
TrialPhaseTreatmentEligibilityNo.Primary EndpointStudy Start
  • NCT indicates National Clinical Trials; FACT, Fulvestrant Combination Therapy; ±, with or without; ET, endocrine therapy; TTP, time to progression; SOFEA, Study Of Faslodex with or without concomitant Arimidex vs Exemestane following progression on nonsteroidal aromatase inhibitors; NSAI, nonsteroidal aromatase inhibitor; PFS, progression-free survival; SWOG, Southwest Oncology Group; HT, hormone therapy.

  • *

    See National Institutes of Health 2008.72

NCT00256698 (FACT)3Anastrozole ± fulvestrantLocal recurrence or metastases; no prior ET for advanced or recurrent disease512TTPJan 2004
NCT00253422 (SOFEA)3Fulvestrant ± anastrozole vs exemestanePrior single-agent NSAI750PFSMar 2004
NCT00075764 (SWOG-S0226)3Anastrozole ± fulvestrantNo prior HT for recurrent or metastatic disease; prior adjuvant HT allowed690TTPApr 2004

Data from preclinical studies and from a few clinical studies indicate that, as discussed above, crosstalk between the ER and growth factor pathways, such as EGF, HER-2/neu, IGF-1, and phosphatidylinositol 3-kinase/Akt, is critical in developing resistance to endocrine therapy.18, 92 Studies also have demonstrated that, in breast tumors, estrogens and ERs can stimulate the enhanced expression or activation of key molecules involved in the regulation of these pathways, which, in turn, can phosphorylate and thus activate ERs in the absence of an estradiol ligand, resulting in loss of estrogen dependence.37, 43 The evidence that enhanced growth factor signal transduction is a major factor in the development of endocrine resistance suggests that combining signal transduction inhibitors with endocrine therapy may delay or supplant the emergence of this acquired resistance.

Many studies are examining the effectiveness and safety of combinations of hormone therapies with signal transduction inhibitors to overcome endocrine resistance by targeting possible crosstalk between the ER and growth factor signaling pathways (Table 3).72, 93–95 Some of these combinations involve tyrosine kinase inhibitors that inhibit EGFR (gefitinib and erlotinib), VEGF receptor (VEGFR) (cediranib), and PKCβ mediated Akt signaling (enzastaurin) as well as multitargeted tyrosine kinase inhibitors, such as sorafenib (inhibits c-Raf, b-Raf, KIT, fms-related tyrosine kinase 2, VEGFR, and platelet-derived growth factor receptor-β) and lapatinib (inhibits EGFR and HER-2/neu). Other combinations involve monoclonal antibodies, such as bevacizumab, which targets VEGF, and trastuzumab, which targets HER-2/neu. Potential trials of c-src kinase inhibitors in combination with hormone therapy also are under consideration.96 Ongoing clinical investigations will determine which of these hypothetical combinations will translate to clinical benefits.

Table 3. Recently Completed and Ongoing Trials of Signal Transduction Inhibitors and Monoclonal Antibodies Combined With Aromatase Inhibitors or Fulvestrant in Postmenopausal Women With Estrogen Receptor-Positive Advanced Breast Cancer That Recurs of Is Refractory to Prior Endocrine Therapy*
TrialPhaseTreatmentEligibilityNo.Primary EndpointStudy Start
  • TAnDEM indicates the Trastuzumab in Dual HER2/Estrogen Receptor-Positive Metastatic Breast Cancer trial; TAM, tamoxifen; HER2, human epidermal growth factor receptor-2; +, positive; PFS, progression-free survival; ESMO, European Society for Medical Oncology; NCT, National Clinical Trials; HT, hormone therapy; CBR, clinical benefit rate; ORR, overall response rate; NSAI, nonsteroidal aromatase inhibitor; MBC, metastatic breast cancer; AI, aromatase inhibitor; HR, hazard ratio; CALGB, Cancer and Leukemia Group B; ±, with or without; PFS, progression-free survival; PK, pharmacokinetics; EORTC, European Organization for Research and Treatment of Cancer; NA, not available.

  • *

    See National Institutes of Health 2008.72

  • See Kaufman 2006.95

TAnDEM Trial2Anastrozole and trastuzumabPrior TAM; HER2+208PFSData presented at ESMO 2006
NCT001792962Letrozole and erlotinibNo more than 1 prior chemotherapy and 1 prior HT150CBRNov 2003
NCT001181572Lapatinib and TAMPrimary or secondary resistance to prior TAM41ORRMay 2005
NCT002382902Trastuzumab and letrozolePrior NSAI40ORRMay 2005
NCT002173991/2Sorafenib and anastrozoleMBC, ≥1 prior AI50CBRJun 2005
NCT002400712Bevacizumab and prior HRResponse and then resistance to prior HT30Safety and efficacyOct 2005
NCT00390455 (CALGB 40302)3Fulvestrant ± lapatinibPrior HT (1 or 2)324PFSSept 2006
NCT004241641Lapatinib and TAMPrior AI, prior TAM20PKNov 2006
NCT004239172Fulvestrant and bevacizumabPrior AI516-Mo PFSAug 2007
NCT004515552Fulvestrant ± enzastaurinResistance to AI132CBRMar 2007
NCT004548052Fulvestrant and AZD2171Prior HT, no fulvestrant; no more than 1 prior chemotherapy64PFSApr 2007
NCT00066378 (EORTC 10021)2Anastrozole ± gefitinibFailure on prior TAM; no prior AI for MBC108PFS at 1 yNA

Clinical investigation of the potential role of ERβ in modulating breast cancer progression is in its initial stages. It has been established that ERβ alters ERα activity, and ERβ and its splice variants, such as ERβ2/ERβcx, may have an impact on the responsiveness of breast tumors to hormone agents. Results of recent retrospective studies suggest that tumor expression of specific isoforms of ERβ is associated with clinical outcome of adjuvant endocrine treatment for primary breast cancer.24 Such work may lead to new approaches in the management of HR-positive breast malignancies in the near future.

Although the optimal sequence of hormone therapies remains to be determined in randomized, controlled clinical trials, effective agents currently exist that prolong life, reduce or postpone the need for cytotoxic chemotherapy, and maintain QOL in patients who have progressed on or become resistant to initial hormone therapy. Experimental evidence suggests the potential utility of sequencing or alternating various agents with different mechanisms of action and proven efficacy and tolerability profiles to extend the viability of hormone therapy in women with metastatic breast cancer. However, clinical trials need to be designed to test these strategies. Ongoing investigation will explore further the clinical utility of available hormone agents in combination with one another and in combination with novel, targeted therapies in patients who fail on initial hormone therapy.


We thank Monica Nicosia and Tara Ruest of Health Learning Systems for providing editorial support in the preparation of this review.