, Volume 67, Issue 16, pp 2335–2353

Estrogen Receptor Modulators and Down Regulators

Optimal Use in Postmenopausal Women With Breast Cancer


  • Christa K. Baumann
    • Clinic and Policlinic for Medical Oncology, Inselspital
    • Clinic and Policlinic for Medical Oncology, Inselspital
    • International Breast Cancer Study Group Coordinating Center
Review Article

DOI: 10.2165/00003495-200767160-00004

Cite this article as:
Baumann, C.K. & Castiglione-Gertsch, M. Drugs (2007) 67: 2335. doi:10.2165/00003495-200767160-00004


Endocrine treatments have been used in breast cancer since 1896, when Beatson reported on the results of oophorectomy for advanced breast cancer. In the second half of the last century, different endocrine-based compounds were developed and, in this review, the role of the selective estrogen receptor modulators (SERMs) and selective estrogen receptor down regulators (SERDs) in the postmenopausal setting are discussed.

Tamoxifen is the most investigated and most widely used representative of these agents, and has been introduced in the advanced disease, in the neoadjuvant and adjuvant setting, and for the prevention of the disease. Its role has been challenged in recent years by the introduction of third-generation aromatase inhibitors that have proven higher activities than tamoxifen with different toxicity patterns.

Several other SERMs have been investigated, but none have been clearly superior to tamoxifen.

SERDs act as pure estrogen antagonists and should compare favourably to tamoxifen. For the time being, they have been used in the treatment of advanced breast cancers and their role in other settings still needs investigation.

The increased use of aromatase inhibitors as first-line endocrine therapy has resulted in new discussions regarding the role that tamoxifen and other SERMs or SERDs may play in breast cancer. The sequencing of endocrine therapies in hormone-sensitive breast cancer remains a very important research issue.

In 1896, Beatson[1] was the first to describe the deprivation of estrogen (by surgical oophorectomy) as a successful therapeutic option in advanced breast cancer. This was the starting point for research on the relationship between hormones, in particular estrogens, and breast cancer growth, despite the fact that estrogens were discovered only years later and the estrogen receptors were described nearly a century later.

Estrogen receptor modulators first, and then down regulators, were developed in the second half of the last century. Their role in the treatment of breast cancer, both in the early stage and in the metastatic setting, has continued undisturbed until recently when it has been challenged by the use of third-generation aromatase inhibitors, which have been investigated in large clinical adjuvant trials but the use of which in this setting has not been widely accepted.

Tamoxifen is the most extensively researched selective estrogen receptor modulator (SERM) and has been used clinically for decades. The efficacy and adverse effects of tamoxifen are very well known. Tamoxifen has both estrogenic agonistic effects and anti-estrogenic antagonistic effects, which are tissue dependent.

A number of additional compounds have been developed in more recent years, which include: (i) other ‘tamoxifen-like’ SERMs (e.g. toremifene, droloxifene and idoxifene); (ii) the ‘fixed ring’ compound SERMs (e.g. raloxifene, arzoxifene and EM-800); and (iii) the selective estrogen receptor down regulators (SERDs), also called ‘pure anti-estrogens’, such as fulvestrant, SR 16234 (TAS 108), ZK 703 and ZK 253.

1. Pharmacological Properties

Estrogens, SERMs and SERDs act upon estrogen receptors in different ways: estrogen acts only agonistically, SERMs act agonistically or antagonistically, and SERDs act only antagonistically. Estrogens and SERDs have a steroid structure, whereas SERMs lack the steroid structure of estrogens but have a tertiary structure that can bind to the estrogen receptor (ER) [see figure 1].
Fig. 1

The chemical structure of selective estrogen receptor modulators and down regulators.

The linkage of either estrogen or a SERM to one of two ER subtypes causes a modification of the configuration of the bonded ER. The bonded complex consists of two ERs and two SERMs (i.e. dimerisation). The newly configured ER results in new interactions with further proteins and with specific DNA sequences. After spontaneous dimerisation, coactivator and corepressor proteins activate or inhibit, directly or indirectly, the transcription and expression of different genes. This process leads to an intrinsic reaction, which has been investigated in different tissues containing ERs, such as bone, endometrium, healthy breast tissue and breast cancer. In contrast to this, tamoxifen acts as an estrogen agonist in endometrial cells and as an estrogen antagonist in breast cells.[2]

There are two different types of ERs (ERα and ERβ), which are present in varying concentrations in target cells. ERα is almost always an activator and ERβ an inhibitor. If ERβ forms a heterodimer with ERα, it inhibits the action of the ERα. The different SERMs bind to both ER isoforms. In general, there is an agonistic effect after binding to the α receptor and the opposite after binding to the β receptor. This is a further reason for the tissue-specific and ligand-specific effect on ER-containing tissues. A further observation is that the ER configuration is unique for every different ligand that can bind to an ER (e.g. estradiol, tamoxifen, raloxifene and ICI 164384).[36]

SERDs (e.g. fulvestrant) have different structures and another mechanism of action when compared with SERMs. Like endogenous estradiol, they have a steroid structure (figure 1) and like a SERM, they bind to an ER. Because of a long, bulky side-chain, the ER dimerisation described earlier in this section is sterically not possible. Therefore, fulvestrant has a much higher ER-binding affinity than tamoxifen.[7] When fulvestrant is bonded to an ER, the bonded receptor is rapidly degraded, which leads to the loss of ER protein. A reduction in cellular levels of ER with fulvestrant treatment has been observed.[8] The whole ER-mediated signal transduction pathway from the cytoplasm to the nucleus is disrupted and all the estrogen-dependent genes are not expressed. This extent of blockade and inactivation of the ER, called ‘down regulation’, is unique with SERDs conferring a pure antiestrogen effect to these drugs. There are hypotheses that ER destabilisation is more pronounced after treatment with the SERD ZK 703, than with fulvestrant after an observation was seen in xenograft models.[9] ZK 703 and ZK 253 are interesting compounds because of the subcutaneous and oral administration.[9] The action of fulvestrant as a pure antiestrogen on both the ERα and the ERβ is different from the effect of SR 16234. SR 16234 is also viewed as a pure antagonist mainly of ERα. It modulates the recruitment of coregulators to ERs and exhibits agonistic properties on ERβ in the absence of estradiol.[10]

Besides efficacy, the adverse effects of SERMs and SERDs on different organs or systems also need to be carefully evaluated.

Bone cells contain ER of the isoform α and β. Tamoxifen has been reported both to increase and to decrease[11] the risk of hip fracture. Most clinical studies have shown that tamoxifen acts as an agonist for bone,[12] but in premenopausal women it has been reported to act as an antagonist leading to a decrease of the bone density of the lumbar spine and the proximal femur.[13] Raloxifene increases bone density quite as effectively as bisphosphonates (alendronate or risedronate). The incidence of vertebral fractures decreases by 30–50%. Hip fracture incidence and other non-vertebral fracture incidences are not significantly influenced by raloxifene.[14]

Unopposed estrogen intake in postmenopausal women with an intact uterus increases the incidence of endometrial carcinoma. The occurrence of this tumour increases by 2.5-fold with tamoxifen. The same observation has been seen with other SERMs (such as toremifene[15]), whereas raloxifene and fulvestrant have not shown this effect.[11,16] Moreover, no increase of postmenopausal bleeding could be observed with raloxifene when compared with placebo.[17] The reason for this difference in the effect is complex and the different chemical structure of raloxifene that binds to the ER plays the main role. Different patterns of ligand-induced responses can result. An estrogen antagonistic region of raloxifene is characterised by a piperidine side chain. The orientation of this side chain may contribute to the lack of uterotrophic effects seen with raloxifene.[18,19]

Fulvestrant neither increases endometrial thickness nor is expected to increase endometrial cancer incidence because it is a pure antiestrogen and lacks a tissue or receptor-dependent estrogenic effect.[20,21]

2. Classification of the Different Compounds

2.1 Selective Estrogen Receptor Modulators (SERMs)

2.1.1 Tamoxifen

Tamoxifen is a non-steroidal triphenylethylene derivative. It acts as an antiestrogen on breast cancer and as an estrogen on bone, blood lipids and the endometrium.[22] It has been used for the treatment of advanced breast cancer for >30 years.[23,24] Tamoxifen was tested and used in the adjuvant setting of early breast cancer in the 1970s and 1980s.[25] In 1985, tamoxifen received FDA approval in the US as an adjuvant treatment for node-positive breast cancer in postmenopausal women receiving chemotherapy. In 1986, tamoxifen alone was FDA approved for the same indication.

2.1.2 Tamoxifen-Like Drugs

In order to improve the efficacy and reduce the toxicity of tamoxifen, researchers tried to change the chemical structure of the drug. One possibility was to alter the side chains in order to produce new ‘tamoxifen-like’ analogues (toremifene, droloxifene, idoxifene), which became known as triphenylethylene derivatives. The only difference between toremifene and tamoxifen is the additional chlorine atom at position 4. Droloxifene has an additional hydroxy molecule at position 3. The additional side chain of idoxifene contains an iodine atom at position 4 and is called pyrrolidine side chain.

2.1.3 Fixed-Ring Drugs

The tamoxifen structure can be changed into a fixed-ring structure by altering the non-steroidal triphenylethylene ring. This results in the benzothiophene derivatives raloxifene and arzoxifene, and in the benzopyrane derivative EM 800. EM 800 is a prodrug of the active benzopyrene derivative EM 652 (acolbifene, SCH 57068).[26]

2.2 Selective Estrogen Receptor Down Regulators (SERDs)

2.2.1 Fulvestrant

Fulvestrant is a steroidal 7α-alkylsulfinyl analogue of estradiol (figure 1). It has a higher affinity for the ER compared with tamoxifen (8.9% vs 2.5% of the binding affinity of estradiol, respectively) and acts as pure antiestrogen.[7] In the 1990s, it was introduced in the human clinical setting and has shown promising response rates in advanced breast cancer in postmenopausal women.[27]

2.2.2 SR 16234 (TAS 108)

SR 16234 is a steroidal compound with a high affinity to the ERα and acts only antagonistically through ERα. This specific affinity to ERα is unique to SR 16234. However, the compound shows a partial agonistic activity on ERβ. In animal studies, it has been shown that SR 16234 does not increase loss of bone density.[28] The ultrasonographic evaluation of the endometrium in clinical trials has shown that it has a minimal uterotrophic effect.[29] So far, it has been investigated in the treatment of advanced breast cancer,[30] and its favourable profile could make it an ideal drug for breast cancer prevention and treatment.

2.2.3 ZK 703 and ZK 253

ZK 703 and ZK 253 are pure antiestrogens that can be administered subcutaneously. ZK 253 is a structurally similar form of ZK 703 and is also orally bioavailable.[9] They inhibit tumour growth more effectively than tamoxifen and fulvestrant in vivo xenograft breast cancer models. Even in tamoxifen-resistant xenograft tumours, ZK 703 and ZK 253 are more effective than fulvestrant. So far, clinical studies have not been published.

Table I summarises the dosage and administration of SERMs and SERDs.
Table I

Dosage and administration of selective estrogen receptor modulators (SERMs) and selective estrogen receptor down regulators (SERDs)

3. Clinical Use of SERMs and SERDs in Postmenopausal Women with Breast Cancer

The majority of breast tumours in postmenopausal women express the ER.[31,32] ER expression is associated with favourable tumour characteristics in two ways: (i) ER positivity is a predictive factor for response to treatment with a SERM or a SERD;[33] and (ii) ER positivity is possibly a marker of better prognosis.[34] Most of the ER+ tumours also express progesterone receptors (PRs) [75%[31] ]. There is a positive correlation between therapeutic effectiveness of endocrine therapies, and ER and PR expression. In the very small group of tumours that are ER− and PR+ (8.9% in a retrospective analysis[35]), there is an indication that the responsiveness to tamoxifen is not as high as in patients with ER+/PR+ tumours and this is particularly true for the subgroup of patients who have tumours overexpressing human epidermal growth factor receptor type 2 (HER2).[36,37] In these patients, aromatase inhibitors could play an especially important role.

An even more frequent and controversial situation is represented by the ER+/PR− breast cancers. It has been observed that breast cancers can evolve during tamoxifen therapy from ER+/PR+ to ER+/PR−.[38] PR is an estrogen-regulated gene product, and its synthesis requires estrogen and ERs. In the process of losing the PR, breast cancer develops resistance to tamoxifen. Moreover, untreated ER+/PR− breast cancer may have a unique behaviour.[39] In the ATAC (Arimidex [anastrozole], Tamoxifen, Alone or in Combination) trial, retrospective subgroup analysis shows a different behaviour of ER+/PR− and ER+/PR+ disease. The hazard ratio (HR) of the recurrence rate for anastrozole versus tamoxifen was 0.48 for ER+/PR− disease and 0.82 for ER+/PR+ disease. This indicates that adjuvant anastrozole was by far more effective than tamoxifen in patients with ER+/PR− tumours.[40] These findings have not been confirmed after central review of ER and PR.[41] In the Breast International Group (BIG) 1–98 study (tamoxifen vs letrozole), a recently presented retrospective subgroup analysis, with its well known methodological limitations, did not confirm the superiority of aromatase inhibitors over tamoxifen in this subpopulation (disease-free survival [DFS] with letrozole vs tamoxifen: HR for ER+/PR+ tumours was 0.84 and for ER+/PR− tumours was 0.83).[42] These different findings remain unexplained and the role of tamoxifen and aromatase inhibitors needs further clarification in these patients.[43,44]

3.1 Advanced Breast Cancer

3.1.1 First-Line Endocrine Therapy

Response rates of first-line tamoxifen are between 20% and 35% in unselected patient populations (30–60% with unknown hormone receptor status). The clinical benefit (i.e. complete remission, partial remission or stable disease for ≥24 weeks) is between 40% and 60% in unselected patients.[45,46] The duration of response is 2–40 months, with a mean of 15 months. In the 1980s and 1990s, some studies compared tamoxifen with diethylstilbestrol and medroxyprogesterone.[47,48] These compounds have shown similar, or even better, response rates and overall survival (OS) rates than tamoxifen, but they are more toxic and their use has mostly been abandoned. Tamoxifen has remained the first-line drug for the treatment of advanced breast cancer for a long time, but its introduction in the adjuvant setting has resulted in the development of new endocrine strategies, such as the SERDs and the aromatase inhibitors, because of resistance to tamoxifen in the advanced setting after use in the adjuvant situation.

The acute and long-term toxicity of tamoxifen is well documented. In 2–13% of patients with bone metastasis, a tumour flare reaction characterised by strong bone pain and symptomatic hypercalcaemia occurs. In a recently described patient group, the flare began between 2 days and 3 weeks after the start of tamoxifen intake, and lasted ≈9 days.[49] A correlation between a strong tumour flare and a better tumour response to tamoxifen after this flare period has been observed.[5052] An explanation for the phenomenon of flare is the possible initial estrogenic effect of tamoxifen on breast cancer.

In recent years, tamoxifen has been compared with third-generation aromatase inhibitors.

A phase III trial of 916 patients with hormone-receptor unknown or positive tumours showed the superiority of letrozole over tamoxifen for: (i) time to progression (TTP) [median 9.4 vs 6.0 months, respectively; p < 0.0001]; (ii) time to treatment failure (median 9 vs 5.7 months, respectively; p < 0.0001); and (iv) overall response rate (32% vs 21%, respectively; p = 0.0002). An OS benefit in the first-line use has not been documented.[46,53]

Two randomised trials that included 1021 patients with advanced breast cancer, either hormone-receptor unknown or positive, compared anastrozole with tamoxifen as first-line therapies.[45] The median TTP was 7 months for tamoxifen and 8.5 months for anastrozole (p = 0.103; 2-sided T-test). In a subgroup of patients with ER+ and/or PR+ disease, the TTP was significantly longer with anastrozole than tamoxifen (10.7 month and 6.4 month, respectively; p = 0.022). The response rate (complete response and partial response) was superimposable for both drugs (29% for anastrozole and 1% for tamoxifen). OS has not been reported. Both treatments were well tolerated. In a Spanish phase III trial with 238 patients, even OS was significantly prolonged with anastrozole compared with tamoxifen (17.4 months vs 16 months; p = 0.003).[54]

Patients receiving anastrozole were reported to have significantly fewer thromboembolic events (3.6% vs 6.5% for anastrozole and tamoxifen, respectively, irrespective of causality; p = 0.043) and fewer vaginal bleeding events. Bone fractures were surprisingly similar in both groups (2.2% with anastrozole and 2.9% with tamoxifen), possibly because of the relatively short duration of the treatments. Vaginal dryness was not significantly more frequent with patients taking anastrozole (1.4% vs 0.6% with tamoxifen).[45]

In several phase II studies of advanced breast cancer, toremifene has produced response rates of 50–70% in dosages of 60–240 mg/day. The 60mg dose of toremifene was shown to be as effective as the 240mg dose and has been retained in further clinical trials and in clinical practice.[55,56] Five phase III trials in 217–648 patients with mainly ER+ tumours or unknown receptor status (one small trial included only patients with ER+ tumours) have compared toremifene with tamoxifen.[5560] Response rates (between 21% and 38%), TTP, OS and even tolerability of toremifene were similar to those of tamoxifen.

The clinical development of droloxifene has been stopped by the pharmaceutical company because of phase III trials showing inferiority compared with tamoxifen in advanced breast cancer.[61,62]

Idoxifene appears to have more antagonistic and less agonistic effects on the ER-containing tissues. In a trial with 219 patients, it was compared with tamoxifen in the first-line therapy of advanced breast cancer. Response rates were 13% for idoxifene and 9% for tamoxifen. The clinical benefit rate (complete response, partial response and stable disease ≥6 months) was 34.3% for idoxifene and 38.7% for tamoxifen (p = 0.31). The median TTP was slightly, but not statistically significantly, higher for idoxifene than for tamoxifen (140 days for tamoxifen and 166 days for idoxifene). Toxicity was similar in the two examined groups.[62] Idoxifene has also not been further developed because it lacks superiority over tamoxifen and there are concerns about uterine pathology.[62,63]

Raloxifene has a favourable toxicity and tolerability profile. As first-line therapy for advanced breast cancer, it was tested in a small trial of 21 patients and showed a partial response in 33% of patients or stable disease for a median duration of 22 months.[64] It has been further developed for the prevention of bone loss in postmenopausal women and is used widely in the therapy of osteoporosis,[65] and in the prevention of breast cancer in high risk women.[6668]

Arzoxifene showed a response rate of 30–40% with minimal toxicity in phase II trials in first-line therapy.[69,70] Laboratory studies have shown that arzoxifene is 1200-fold more potent than tamoxifen in inhibiting the growth of breast cancer cells and that it does not appear to promote endometrial or uterine growth. A phase III study comparing arzoxifene with tamoxifen in the first-line setting was initiated,[71] but the trial was stopped by the pharmaceutical company developing it after the findings from the first interim analysis suggested that arzoxifene was inferior to tamoxifen with regard to TTP. The data have not yet been published and no further trials are planned to evaluate arzoxifene in advanced breast cancer patients. However, the compound is currently under evaluation for the treatment osteoporosis in postmenopausal women.

EM 800, the active metabolite of which is acolbifene, is an atypical SERM with pure antiestrogenic effects, which could play a role in the therapy of breast cancer. A phase III trial in the first-line setting comparing acolbifene with anastrozole has been planned.[72]

New alternatives to the SERMs are arising with the use of SERDs.

Fulvestrant has been compared with tamoxifen in a trial of 587 patients with advanced breast cancer and positive or unknown hormone receptors.[73] The response rates of tamoxifen and fulvestrant were similar (33.9% vs 31.6%, respectively; p = 0.45), but it could be shown that the clinical benefit (complete response, partial response and stable disease >24 months) was significantly better for tamoxifen than for fulvestrant (62% vs 54.3%, respectively; p = 0.026). At a median follow-up of 14.5 months, there was a non-significant difference favouring tamoxifen with respect to the primary endpoint of TTP (6.8 months for fulvestrant and 8.3 months for tamoxifen; p = 0.088). In an unplanned, retrospective, subgroup analysis of the patients with known ER+ and/or PR+ tumours, the median TTP was 8.2 months for fulvestrant and 8.3 months for tamoxifen (p = 0.39).[73]

The tolerability has been shown to be very similar for the two treatments.[73] Further evaluation will be necessary in order to define the most appropriate patient group benefiting from fulvestrant in the first-line setting.

3.1.2 Second- and Multiple-Line Endocrine Therapy

Tamoxifen was been used in second- and multiple-line treatment of metastatic breast cancer at the beginning of its clinical development.

A more recent trial investigated the ideal sequence of endocrine therapies in the treatment of metastatic breast cancer. Patients were treated with anastrozole or tamoxifen as first-line therapies and were switched after progression to the alternative compound. Median TTP from treatment crossover was 6.7 months for tamoxifen (after anastrozole [n = 19]) and 5.7 months for anastrozole (after tamoxifen [n = 18]). TTP from entry into the trial to the first tumour progression was 11.3 months for first-line anastrozole (n = 31) and 8.3 months for first-line tamoxifen (n = 29) [p = 0.75]. The anastrozole-tamoxifen sequence had a time from randomisation to second progression of 2 months and an OS of 69.7 months; the same values for the tamoxifen-anastrozole sequence were 19.5 months and 59.3 months, respectively. OS was not significantly longer for the anastrozole-tamoxifen sequence (p = 0.10). This trial suggests that tamoxifen is an effective second-line endocrine therapy after anastrozole.[74]

In recent years, the use of tamoxifen has shifted to the adjuvant setting and its relevance in second- and multiple-line therapies has dramatically decreased. The superiority of aromatase inhibitors in the adjuvant treatment may re-establish a more important role for tamoxifen in the metastatic setting.

Early or late tamoxifen treatment results in resistance to the compound and consequent progression of the disease. Interestingly, tamoxifen-resistant breast cancers can remain estrogen dependent and about half of them respond to a second-line endocrine therapy.

There are several hypotheses about the reasons for tamoxifen resistance. Some studies postulate ERα mutations, some report the changes in the expression of ERα or ERβ, or the alterations in co-regulatory proteins and the influences of cellular kinase signal transduction pathways. It is interesting to observe that after prolonged tamoxifen use, not only does resistance develop, but breast cancers even seem to be stimulated by tamoxifen as well as by toremifene and raloxifene.[7577] Some of these tumours regress after withdrawal of the compound.[78,79] It has also been observed that a conversion of resistance can occur by treating patients with an aromatase inhibitor or a SERD. In a second phase of resistance, it has been observed that estradiol inhibited tumour growth but SERMs still stimulated it. The cancer cells learn to grow without estrogen and a treatment with estrogen at this point induces apoptosis in these cancer cells again.[80]

The new generation SERMs, such as toremifene, idoxifene, raloxifene and arzoxifene, have shown a high level of cross resistance with tamoxifen[55,59,64,69,81] and are therefore inadequate as second-line treatments after tamoxifen failure. Toremifene has shown response rates between 0% and 14% in more advanced phases of the disease[52] and raloxifene has shown, in a small trial with 14 tamoxifen-resistant patients, only one minor response.[82]

The non-steroidal EM 800 and its active metabolite (acolbifene) are orally bioavailable and have recently entered into clinical investigation. A phase II study with 43 postmenopausal women with advanced breast cancer, randomised in two groups receiving different doses (20 or 40 mg/day orally) as second-line therapy (after tamoxifen failure) reported an objective tumour response rate of 10% and 14%, respectively. The drug was well tolerated with both doses and it did not show estrogenic activity on the breast and endometrial tissues.[83]

A phase III trial in which EM 800 in two doses (20 or 40 mg/day orally) was compared with anastrozole was apparently interrupted after some patients were entered because of a lack of superior effectiveness of EM 800 over anastrozole.[84]

The other fixed-ring SERM, arzoxifene, did not show effectiveness in tamoxifen-resistant patients (complete response and partial response 10%).[69]

Two large, multicentre, randomised trials including >800 patients have shown similar outcomes for fulvestrant when compared with anastrozole in postmenopausal women with hormone-receptor positive (ER+ and/or PR+) advanced breast cancer after progression on prior endocrine treatment.[85] The first study included 451 patients, and the median TTP was 5.5 months for fulvestrant and 5.1 months for anastrozole. Response rates were 20.7% for fulvestrant and 15.7% for anastrozole.[86] The second study was conducted in 400 patients with similar results.[87] The tolerability was good and similar in the two groups except for joint disorders and thromboembolic disease, which were more frequent in the anastrozole group (joint disorders were 9.3% with fulvestrant and 13.5% with anastrozole; thromboembolic disease was 3.4% with fulvestrant and 6.7% with anastrozole). The first results of a phase III trial with 693 patients comparing fulvestrant with exemestane after prior non-steroidal aromatase inhibitor therapy have been presented and are available in abstract form.[88] They show a similar TTP (3.7 months in both groups). The duration of response for responding patients was 13.5 months versus 9.8 months for fulvestrant and exemestane, respectively.[88]

A higher response rate was expected from the xenograft models, especially in tamoxifen-resistant tumours. In these clinical trials, only ≈20% of tamoxifen-resistant tumours responded to fulvestrant. A possible reason for this low response rate may be the lower plasma concentration that can be achieved in patients by the intra-muscularly administered fulvestrant compared with the concentrations achieved in cell cultures. It would be interesting to investigate orally and subcutaneously applicable SERDs such as ZK 703 and ZK 253.

In a phase II study, fulvestrant showed efficacy after a therapy with third-generation aromatase inhibitors; 35% of patients had a clinical benefit (14.3% showed partial responses and 20.8% stable disease).[89] The efficacy in this setting is also supported by data from the fulvestrant compassionate-use programme with similar response rates in heavily pretreated women. Most of them had received an aromatase inhibitor prior to fulvestrant.[90]

SR 16234 acts as a pure antagonist of the ERα. A phase I study has shown that SR 16234 has anti-tumoural activity and is well tolerated. With different doses from 40 to 160 mg/day orally, there are only grade 1–2 toxicities (mainly headache, hot flashes, nausea, vomiting and sweating).[29] Phase II and III studies are being planned.[30]

Further studies will have to determine which sequence of hormonal therapies is optimal in the advanced setting considering the shift of the use of aromatase inhibitors to the adjuvant setting.

3.2 Neoadjuvant Treatment of Breast Cancer

In early breast cancer, studies comparing endocrine therapy with chemotherapy have shown a longer DFS and OS for patients with hormone-responsive disease when treated by endocrine therapy.[9193]

Tamoxifen has been analysed in elderly patients with locally advanced or operable breast cancer instead of primary surgery. These trials have shown that immediate surgery confers improved local control compared with tamoxifen, but patients treated with initial tamoxifen had fewer metastatic relapses and a similar survival time.[9497] A phase III study that compared tamoxifen with multimodal treatment in pre- and postmenopausal patients showed that patients taking tamoxifen alone had a significantly shorter first locoregional failure time than patients treated with the multimodal concept (including neoadjuvant chemotherapy, radical mastectomy and adjuvant hormone therapy). However, after a follow-up of 52 months, there was no statistically significant difference in the OS and DFS between the two groups. It is mentionable that in this study patients were included regardless of the hormone receptor status of their tumours.[98]

Tumour size reduction has been reported from a trial including 324 patients; 46% of patients treated with tamoxifen had a tumour volume reduction of >50%, but 88% of patients treated with letrozole and 78% of patients treated with anastrozole reached a significant tumour reduction.[99,100] All of these trials compared endocrine treatment alone with surgical treatment and were therefore not primarily investigating neoadjuvant therapy (which should be followed by surgery).

The IMPACT (IMmediate Preoperative Arimidex, Tamoxifen, or Combined with Tamoxifen) trial compared anastrozole with tamoxifen, or a combination of both, as preoperative treatment of ER+ operable breast cancer in 330 postmenopausal women who were treated for 3 months.[101,102] Anastrozole was superior to tamoxifen for the 124 patients who needed a down staging to achieve breast conserving surgery because of the initial need for mastectomy (44% of the patients after anastrozole and 31% after tamoxifen [p = 0.23] could undergo breast conserving surgery).

The PROACT (Pre-Operative Arimidex Compared to Tamoxifen) trial conducted in 457 patients confirmed these results.[103]

Results of trials with letrozole versus tamoxifen are comparable in terms of the goal of breast conserving surgery. However, the clinical response rate for letrozole is superior to that of tamoxifen (55% vs 36%; p < 0.001).[104] In contrast, the clinical overall response rate for anastrozole compared with tamoxifen was not as favourable (anastrozole 37% vs tamoxifen 36%).[101,102]

A hypothetical problem with using tamoxifen in the neoadjuvant setting is the long time (up to 5 weeks) required to reach a therapeutic and steady-state plasma concentration. In contrast, aromatase inhibitors reach a therapeutic plasma concentration more quickly (within days).[99]

There are few data available on the use of fulvestrant in the neoadjuvant setting. In one trial, 201 patients were randomised to receive intramuscular fulvestrant in doses of 50, 125 or 250mg monthly or oral tamoxifen in the usual dose of 20 mg/day, or matching placebo for 14–21 days before surgery. The aim of this study was primarily to make observations about tumour reaction. The nuclear ER content was significantly reduced in tumours of patients who were treated with fulvestrant 250mg. Tamoxifen reduced ER content similarly to the two lower doses of fulvestrant (50mg and 125mg). All the doses of fulvestrant produced a reduction of Ki-67.[8]

3.3 Adjuvant Treatment of Breast Cancer

Tamoxifen was the first endocrine active compound used for early breast cancer in the adjuvant setting.

The Nolvadex Adjuvant Trial Organisation (NATO) identified tamoxifen as a drug that delayed the recurrence of breast cancer.[105] Patients were randomised to tamoxifen 10mg twice daily for 2 years versus no further therapy after curative local surgery. After a mean follow-up of 21 months, significantly fewer events, defined as first recurrence of breast cancer and death without confirmed recurrence, occurred in the tamoxifen group compared with the control group (14.2% vs 20.5%; p = 0.01).

In addition, in the Stockholm trial,[106] which investigated node-negative breast cancer in postmenopausal women, patients received tamoxifen (40mg per day) for 2 years or no adjuvant hormonal therapy. Patients in the tamoxifen arm who were disease free after 2 years were randomised in two further groups. One discontinued treatment, the other received tamoxifen for 3 more years. After a median follow-up of 7 years, the tamoxifen-receiving group experienced a significant prolongation of DFS compared with the control group (p < 0.01) and a reduction of deaths (p = 0.02). Survival was longer, but not statistically significantly (p = 0.11). Tamoxifen benefit was observed only in patients with ER+ breast cancer. In a further follow-up, the event-free survival and the OS became statistically significant for the 5-year tamoxifen group.[107]

The Scottish Trialists’ Group conducted a trial investigating the ideal duration of tamoxifen treatment and randomised 1312 patients to tamoxifen for 5 years or tamoxifen at first relapse.[108110] At a median follow-up of 47 months, 157 patients (24%) in the 5-years tamoxifen patient group and 250 patients (38%) in the other patient group developed recurrent disease. The delay in recurrence was independent of nodal status. DFS and OS were statistically significantly longer for the 5-year tamoxifen patient group compared with the tamoxifen-at-relapse patient group (p < 0.0001 for DFS and p = 0.002 for OS). This trial and a further confirmatory trial clearly defined the benefit of tamoxifen when given in the adjuvant situation for 5 years, but a longer treatment with this compound has not shown additional benefits.[108,109] The National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14 trial evaluated patients with ER+, node-negative cancer who had been given tamoxifen either for 5 years or placebo for 5 years.[111] An additional randomisation was conducted to define if 5 additional years of tamoxifen yielded even better results.[112] After a follow-up period of 10 years, OS was significantly shorter in patients who received longer treatment.

The Oxford overview[113] compiled the results from 37 000 women included in 55 trials comparing the effect of tamoxifen with no further endocrine treatment. The annual recurrence rate for ER+ breast cancer was almost halved and the breast cancer mortality rate was reduced by a third. These positive results for tamoxifen were independent of age, node status and the use of chemotherapy.

The acute- and long-term toxicity of tamoxifen is well documented in several adjuvant trials. In the Oxford overview,[113] 2% of patients experienced vaginal bleeding, 9% thromboembolic events (5–6 per 1000 patient-years) and 0.7% developed an endometrial cancer.

The question about the sequence of treatment modalities in the adjuvant setting is of interest. Patients offered antiestrogen therapy after completion of chemotherapy experienced less risk of recurrence compared with the concomitant administration of the two modalities.[114]

As an alternative to tamoxifen, with the aim of reducing adverse effects, toremifene has been investigated in International Breast Cancer Study Group (IBCSG) trials since 1993. IBCSG trial 12-93 investigated peri- and postmenopausal patients with node-positive, ER+ breast cancer who were considered suitable for endocrine therapy alone, and IBCSG trial 14-93 investigated patients who required chemotherapy prior to endocrine treatment. Patients were randomised to receive either tamoxifen 20 mg/day or toremifene 60 mg/day. A total of 1035 patients were evaluable, 81% had received prior chemotherapy. Toremifene and tamoxifen yielded similar DFS and OS rates after a median follow-up of 5.5 years and toxicity, in particular endometrial cancer incidence, was low for both treatments (six in the toremifene group [1%], and two in the tamoxifen group [<1%]).[15]

Tamoxifen has been compared with aromatase inhibitors up front in two large adjuvant trials (ATAC:[115] tamoxifen vs anastrozole; BIG 1-98:[116] tamoxifen vs letrozole) and the results have shown a superiority of the aromatase inhibitors over tamoxifen in terms of DFS. In the ATAC trial,[115] 3125 patients were randomised to anastrozole, 3116 patients to tamoxifen and 3125 patients to the combination of both. After a follow-up period of 4 years, DFS was significantly higher for the anastrozole group than for those including tamoxifen alone (86.9% vs 84.5%; HR 0.86 in favour of anastrozole; p = 0.03). After the completion of 5 years of adjuvant treatment, anastrozole led to even more significant improvements compared with tamoxifen alone (575 vs 651 events; p = 0.01).[115] An OS benefit has not been reported. In the follow-up report at 4 years, the safety analysis showed that endometrial cancer (p = 0.007), vaginal bleeding and discharge (p < 0.001), cerebrovascular events (p < 0.001), venous thromboembolic events (p < 0.001) and hot flashes (p < 0.001) occurred less frequently in the anastrozole group. Musculoskeletal disorders and fractures (p < 0.001) occurred less frequently in the tamoxifen-only group.[37,117]

In the BIG 1-98 trial,[118] tamoxifen was compared with letrozole in four treatment arms; (i) 5-years tamoxifen; (ii) 5-years letrozole; (iii) letrozole followed by tamoxifen up to 5 years; and (iv) tamoxifen followed by letrozole up to 5 years. A total of 4001 patients in the letrozole-only group and 4007 patients in the tamoxifen-only group were analysed. After a median follow-up of 25.8 months, 351 events had occurred in the letrozole group and 4 in the tamoxifen group. Compared with tamoxifen, letrozole significantly reduced the risk of an event ending a period of DFS (HR 0.81; p = 0.003), especially the risk of distant recurrence (HR 0.73; p = 0.001). These results have been confirmed in the update at 51 months median follow-up (352 events among the 2463 women receiving letrozole and 418 events among the 2459 women receiving tamoxifen [hazard ratio 0.82; p = 0.007]).[116] The safety profile was similar to the one in the ATAC trial. Women receiving letrozole had a higher incidence of skeletal and cardiac events and of hypercholesterolaemia.[118]

Tamoxifen has been compared with aromatase inhibitors in patients remaining disease-free after 2–3 years of tamoxifen (IES [Intergroup Exemestane Study] with exemestane, ABCSG/ARNO [Austrian Breast Cancer Study Group/German Adjuvant Breast Cancer Group] trial with anastrozole, BIG 1-98 with letrozole [data not yet available] and in a small trial by the GROCTA [Italian Breast Cancer Adjuvant Study Group]) and, in all these trials again, patients who received aromatase inhibitors reached a longer DFS.[119121] In the IES trial, after tamoxifen for 2–3 years, 4742 patients were randomised to continue tamoxifen or to switch to exemestane. After a median follow-up of 55.7 months, 357 events in the exemestane group and 455 events in the tamoxifen group were reported (HR in favour of exemestane was 0.76; p < 0.0001). A trend to improved OS was observed (222 deaths in the exemestane group vs 261 deaths in the tamoxifen group; p = 0.08). OS was not significantly different in the two groups.[121] Because of nearly identical inclusion criteria, patients of the ARNO 95 trial and of the ABCSG trial 8 could be analysed together. Postmenopausal women with hormone-receptor positive (ER+ and/or PR+) early breast cancer who had completed 2 years of adjuvant tamoxifen were included. They received either further tamoxifen (n = 1606) or anastrozole (n = 1618). At a median follow-up of 28 months, 67 events with anastrozole and 110 events with tamoxifen occurred (HR in favour of anastrozole 0.60; p = 0.0009). There were significantly more fractures (p = 0.015) and significantly fewer thromboses (p = 0.034) in patients treated with anastrozole compared with those receiving tamoxifen.

The toxicity profile of tamoxifen and of the aromatase inhibitors is different, and tamoxifen may remain the therapy of choice for some subgroups of patients (e.g. those with osteoporosis or with cardiovascular pathology) despite the lower efficacy.

3.4 Breast Cancer Prevention

Breast cancer remains the most frequent tumour diagnosed (incidence between 120/100 000 and >400/100 000 depending on age and race) in women and one of the major killers (14% of cancer deaths).[122] Therefore, attempts to prevent the disease have started in recent years, including lifestyle modification (e.g. diet, alcohol intake, optimising weight, decreasing exposure to exogenous and endogenous estrogens), ablative surgery (prophylactic mastectomy, oophorectomy), chemoprevention (e.g. retinoids, cyclo-oxygenase-2 inhibitors) and endocrine interventions with SERMs.

The role of estrogen in breast cancer growth has been well established in humans and therefore compounds interfering with this hormone may play a role in the prevention of the disease.

Women with early stage breast cancer treated by tamoxifen had a statistically significantly lower incidence of contralateral breast cancer and a lower risk of recurrence, and they had an improved survival rate compared with women who received placebo.[11,113,123128] Especially mentionable are the long-term results of IBIS (International Breast Intervention Study)-I, which compared 5 years of tamoxifen with placebo. After a follow-up of 96 months, tamoxifen showed a risk-reducing effect without increasing the adverse effects of tamoxifen reported after 5 years; 142 breast cancers were diagnosed in the 3579 women in the tamoxifen group and 195 in the 3575 women in the placebo group (4.97 vs 6.82 per 1000 woman-years; p = 0.004).[128] Therefore, tamoxifen was considered the best candidate for prevention despite its toxicity.

The NSABP-P1 study included a group of women at higher risk of developing breast cancer. The high risk was defined in the inclusion criteria as: (i) >60 years old; (ii) between 35 and 59 years of age with a 5-year predicted risk for breast cancer of at least 1.66%; or (iii) with a history of lobular carcinoma in situ (LCIS) or atypical hyperplasia. Specifically, one quarter of the women had a 5-year predicted breast cancer risk of ≤2%, 58% had a 5-year predicted risk between 2% and 5%, and 17% had a 5-year predicted risk of >5%. The result of the trial showed that after 7 years of follow-up, the cumulative rate of invasive breast cancer was reduced from 42.5 per 1000 women in the placebo group to 24.8 per 1000 in the tamoxifen group (p < 0.001). The cumulative rate of non-invasive breast cancer was reduced from 15.8 per 1000 in the placebo group to 10.2 per 1000 women in the tamoxifen group (p = 0.008). Tamoxifen administration resulted in a most striking reduction in the rate of ER+ invasive breast cancer of 62%, but did not reduce the rate of ER− breast cancer. The risk of endometrial cancer was increased (risk ratio [RR] 3.8) as well as the risk of thromboembolic disease (stroke RR = 1.42, pulmonary embolism RR = 2.15, deep-vein thrombosis RR = 1.44).[11,124]

A meta-analysis of several prevention trials, which were different in size, methods and eligibility of participants, showed that the relative risk reduction of breast cancer was ≈40% when taking tamoxifen.[129]

Two further trials (the Royal Marsden Hospital trial[125] and the Italian Tamoxifen Prevention National Trial[126]) with the same main question did not show a statistically significant overall risk reduction for tamoxifen, but only a trend toward a lower risk in the tamoxifen group. In the Royal Marsden Hospital trial, a statistically significant reduction in the incidence of ER+ breast cancer was reported in the tamoxifen arm after a follow-up of 20 years.[130] During the 8-year treatment, the risk of ER+ cancer was not statistically significantly lower in the tamoxifen arm. Several factors could have contributed to these different results, which include: (i) sample size of the trials; (ii) breast cancer risk characteristics such as the proportion of women at risk for developing ER+ breast cancer; (iii) incidence rates of all breast cancers in the different places; and (iv) concomitant use of hormone therapy.

The fixed-ring SERMs promise an excellent potential for breast cancer prevention treatment because they appear to be devoid of any agonistic activity in the endometrium, while they appear to be potent antiestrogens in the breast and estrogens in the bone. Raloxifene is the most extensively studied SERM in this class. During the development of raloxifene for osteoporosis, it was found to significantly reduce the incidence of breast cancer in postmenopausal women without effects on the endometrium.[14,17,65,131,132]

The superiority of raloxifene over placebo in preventing breast cancer was reported in the Raloxifene Use for The Heart (RUTH) trial.[133] This trial was conducted to investigate the possible reduction of the coronary risk (not observed).

In the Multiple Outcomes Raloxifene Evaluation (MORE) study, raloxifene was tested in two oral dosages of 60 or 120 mg/day. The control group received placebo. The primary endpoint was the development of fractures.[134] A secondary endpoint was the incidence of invasive breast cancer.[17] Among older women with osteoporosis, the risk of ER+ invasive breast cancer was decreased by 90% and the risk of invasive breast cancer was decreased by 76% during 3 years of treatment with raloxifene. Like tamoxifen, raloxifene increased the risk of thromboembolic disease, but did not appear to increase the risk of endometrial cancer.

The Continuing Outcomes Relevant to Evista (raloxifene) [CORE] trial[135] examined the effect of 4 additional years of raloxifene therapy on the incidence of invasive breast cancer in women participating in the MORE trial. In 5213 women after 4 years, the risk of invasive breast cancer was reduced by 69% in the raloxifene group compared with the placebo group.[135] An increase in the endometrial cancer risk did not occur.

The NSABP STAR (Study of Tamoxifen and Raloxifene)[136] compared double-blinded tamoxifen 20 mg/day over 5 years with raloxifene 60 mg/day over 5 years. A total of 19 747 postmenopausal women with a 5-year predicted breast cancer risk of at least 1.66% as determined by the modified Gail Model[137] or a history of LCIS of the breast treated by excision alone, were enrolled. The investigated group of women showed a risk of developing breast cancer with an average Gail Score of >4%. There was no statistically significant difference (p = 0.83) in the incidence of breast cancer between the two drugs. Death from any cause, other invasive cancers, ischaemic heart disease and osteoporotic fractures were similar in the two groups. There was a statistically significant 30% reduction in the risk of thromboembolic events in the raloxifene group compared with the tamoxifen group. Raloxifene was not as effective as tamoxifen in the prevention of non-invasive breast cancer. There were 57 incident cases of non-invasive breast cancer among women in the tamoxifen arm and 80 cases in the raloxifene group (cumulative incidence over 6 years, 8.1 per 1000 in the tamoxifen group and 11.6 in the raloxifene group; p = 0.052). Surprisingly, both compounds showed a similar incidence of endometrial cancers (annual incidence rates of 2 per 1000 with tamoxifen and 1.25 per 1000 with raloxifene; cumulative incidence rates over 7 years, 14.7 per 1000 with tamoxifen and 8.1 per 1000 with raloxifene; p = 0.07). The number of hysterectomies performed during the trial for non cancer-related causes was more than twice as high in the tamoxifen group (244 compared with 111 in the raloxifene group).[136] However, more than half of the women in the trial had a prior hysterectomy (51.6%).

Arzoxifene, with its interesting profile, exceptionally high affinity for the ER compared with other SERMs and lack of estrogenic effect on the endometrium, while maintaining beneficial estrogen-agonist effects on bone an lipids, may also develop as a very interesting prophylactic agent.[71]

The superiority of aromatase inhibitors in the adjuvant setting has stimulated the investigation of these compounds for the prevention of the disease and trials are currently ongoing. IBIS-II[138] and the National Cancer Institute of Canada Clinical Trials Group (MAP.3 trial)[139] are comparing the efficacy of the aromatase inhibitors, anastrozole and exemestane with tamoxifen. Until one of these trials shows a greater benefit in breast cancer prevention, tamoxifen or raloxifene remain the only proven endocrine preventive treatments in postmenopausal women.

4. Conclusion

Tamoxifen is the prime example of the SERMs and has played a striking role in the endocrine treatment of hormone-receptor positive breast cancer in all disease stages for decades, in both post- and premenopausal women. Other SERMs have, in general, failed to show significant advantages over tamoxifen and are generally not widely used.

SERDs have been investigated only in the advanced or neoadjuvant setting and need further evaluation before entering into the adjuvant situation.

In the treatment of advanced breast cancer, different, non cross-resistant endocrine therapies are generally required in a sequential manner as patients relapsing or progressing under the first endocrine therapy generally remain hormone sensitive. More and more aromatase inhibitors are being used as first-line therapy of advanced breast cancer because of their superiority in terms of outcome. We have some data about the effectiveness of a treatment with SERMs and SERDs as a second-line therapy after aromatase inhibitors. However, fulvestrant has been shown to be active after a therapy with an aromatase inhibitor. Tamoxifen can be an effective second-line therapy after anastrozole. The ideal sequence of endocrine therapies in advanced breast cancer still has to be defined.

It also remains to be investigated whether the new orally bioavailable pure antiestrogens acolbifene and ZK 703 will have equivalent or superior efficacy compared with intramuscularly injected fulvestrant.

The role of SERMs in the adjuvant setting will be redefined as in this indication, aromatase inhibitors are probably better alternatives in terms of outcome, but due to their different toxicity profiles, they may be prohibitive for certain patients. For these patients, tamoxifen may remain the first choice in adjuvant treatment.

In the prevention of the disease, raloxifene is possibly superior to tamoxifen, at least with regard to its toxicity profile. Aromatase inhibitors are currently being tested in ongoing trials in this setting. Until these studies are reported, the use of tamoxifen and raloxifene will remain the standard treatment in this indication.


No sources of funding were used in the preparation of this review. The authors have no conflicts of interest that are directly relevant to the content of this review.

Copyright information

© Adis Data Information 2007