Endocrine Manipulation in Advanced Breast Cancer: Recent Advances with SERM Therapies

Home , Droloxifene, Fulvestrant, Idoxifene, Triphenylethylene

4376s Vol. 7, 4376s-4387s, December 2001 (Suppl.) Clinical Cancer Research

Stephen R. D. Johnston e gynecological side effects may prove more beneficial than Department of Medicine, Royal Marsden Hospital and Institute of either tamoxifen or AI. The issue is whether the current Cancer Research, London SW3 6JJ, United Kingdom clinical data for SERMs in advanced breast cancer are sufficiently strong to encourage that further development.

Abstract Tamoxifen is one of the most effective treatments for Introduction breast cancer through its ability to antagonize estrogen- Ever since evidence emerged that human breast carcinomas dependent growth by binding estrogen receptors (ERs) and may be associated with estrogen, attempts have been made to inhibiting breast epithelial cell proliferation. However, ta- block or inhibit estrogen's biological effects as a therapeutic moxifen has estrogenic agonist effects in other tissues such strategy for women with breast cancer. Estrogen has important as bone and endometrium because of liganded ER-activating physiological effects on the growth and functioning of hormone- target genes in these different cell types. Several novel an- dependent reproductive tissues, including normal breast epithe- tiestrogen compounds have been developed that are also lium, uterus, vagina, and ovaries, as well as on the preservation selective ER modulators (SERMs) but that have a reduced of bone mineral density and reducing the risk of osteoporosis, agonist profile on breast and gynecological tissues. These the protection the cardiovascular system by reducing cholesterol SERMs offer the potential for enhanced efficacy and re- levels, and the modulation of cognitive function and behavior. duced toxicity compared with tamoxifen. In advanced breast Thus, a strategy to block or reduce estrogen function in an cancer clinical data exist for three first-generation SERMs attempt to treat or prevent breast cancer could have a severe (toremifene, droloxifene, idoxifene), which are related to the impact on a woman's health by interfering with normal estro- triphenylethylene structure of tamoxifen. In Phase II trials gen-regulated tissues. in a total of 263 patients resistant to tamoxifen, the median For over 30 years, synthetic antiestrogens have been de- objective response rate to these SERMs was only 5% (range, veloped as treatment for ER+ve 3 breast cancer, and the tri- 0-15%), with stable disease for >--6 months in an additional phenylethylene tamoxifen was the first hormonal compound 18% (range, 9-23%). As first-line therapy for advanced with efficacy in advanced breast cancer (1). Tamoxifen is the breast cancer, the median response rate was 31% (range, most widely used and tested drug in breast cancer and is now 20-68%) with a median time to progression of 7 months. recognized to significantly improve survival as adjuvant therapy Randomized Phase III trials for toremifene and idoxifene in in early breast cancer (2), as well as to reduce the incidence of more than 1500 patients showed no significant difference breast cancer in healthy women at risk of the disease (3). compared with tamoxifen. Fewer clinical data exist for the Despite concerns about unfavorable antiestrogenic effects on structurally distinct second- and third-generation SERMs healthy tissues, paradoxically it was discovered that tamoxifen (raloxifene, arzoxifene, EM-800, and ERA-923), although a acted as an estrogen on bone, blood lipids, and the endometrium similarly low median response rate of 6% (range, 0-14%) (4). More recently, the molecular structure and function of ER was seen in Phase II trials in tamoxifen-resistant patients. It biology has been elucidated, revealing how tamoxifen and re- remains unclear whether any clinical advantage exists for lated drugs act as ligands to differentially switch on or off gene second- and third-generation SERMs over tamoxifen as expression in specific tissues (5, 6). The ability of antiestrogens first-line therapy. With the emergence of potent aromatase to have alternative effects on various estrogen-regulated targets inhibitors (AIs) that are superior to tamoxifen, the clinical led to the term "SERM" to describe this class of drugs. It is now questions in advanced disease have shifted to which anties- possible to develop SERMs that range from full estrogen ago- trogen (including SERMs) may be effective following failure nists to pure antagonists with different effects in separate target of AIs, and whether any merit exists for combined AI/SERM tissues. In this way, SERMs offer the potential to treat and therapy. The main advantage for SERM therapy probably prevent a number of conditions ranging from osteoporosis, remains in early stage-disease (adjuvant therapy or preven- menopausal symptoms, cardiovascular disease, and breast and tion), in which the estrogenic effects on bone and reduced endometrial cancer. This article will review the development of SERMs in breast cancer, addressing in particular the limitations of tamoxifen that SERMs have attempted to overcome and the clinical data available to date for each of the SERM compounds. i Presented at the First International Conference on Recent Advances and Future Directions in Endocrine Therapy for Breast Cancer, June 21-23, 2001, Cambridge, MA. 2 To whom requests for reprints should be addressed, at Department of Medicine, Royal Marsden Hospital and Institute of Cancer Research, London, SW3 6JJ, UK. Phone: 44-(0)207-808-2748; Fax: 44-(0)207- 3 The abbreviations used are: ER, estrogen receptor; ER+ve, ER posi- 352-5441; E-mail: [email protected]. tive; SERM, selective ER modulator; CI, confidence interval.

Tamoxifen: The Prototype SERM AF1 AF2 The triphenylethylene nonsteroidal antiestrogen tamoxifen CoRp, was first synthesized in the 1960s and was found to have clinical §247 activity in postmenopausal women with advanced breast cancer

(1). Tamoxifen antagonizes the effects of estrogen in breast | §247 cancer cells by binding ER, inducing G~ cell cycle arrest, and Estradiol inhibiting tumor growth. Although tamoxifen prevented growth of ER+ve breast tumor xenografts in vivo (7), at the same time it stimulated uterine growth (8) and supported the growth of endometrial xenografts in vivo (9). A similar spectrum of estro- CoR genic and antiestrogenic effects emerged in patients, with vag- § inal dryness and hot flushes the most frequently reported antiestrogenic toxicities (10). Because of estrogenic activity of Tamoxifen - ~. Idoxifene tamoxifen in the liver, total serum cholesterol levels were re- Raloxifene ...... duced by 10-15% (11). Likewise, bone mineral density was EM-800 ERE - preserved in tamoxifen-treated postmenopausal women, although in premenopausal women, this effect was not apparent CoR A CoA (12, 13). In patients, tamoxifen functioned as an estrogen on the endometrium, with endometrial thickening and hyperplasia together with an increased risk of endometrial cancer (14). Al- Fulvestrant though some of these additional properties of tamoxifen are of potential benefit for women, SERMs were developed for breast ERE cancer with the aim of reducing some of tamoxifen's toxicities, Fig. 1 Concept diagram to explain spectrum of Ez/tamoxifen/SERM in particular eliminating any agonist effects on the gynecolog- activity through altered structure-function relationships between ligand, ical tract. In terms of breast cancer therapy, a meta-analysis of ER, and the coactivator/corepressor complex that activates or inhibits all clinical trials found that 5 years of tamoxifen in women with estrogen-regulated gene transcription. Representative SERMs are early stage ER+ve breast cancer significantly reduced the risk ranked on the left in order of increasing antagonistic effects. ERE, of recurrence (47% reduction in annual odds) and death (26% estrogen response elements; CoR, corepressor; CoA, coactivator; H12, helix 12 of ER; +/-, presence/absence of transcriptional activity at the reduction in annual odds; Ref. 2). This benefit was greatest in AF-1 and AF-2 sites. women with ER-rich tumors and occurred across all age groups, irrespective of nodal involvement. In addition, tamoxifen's antiestrogenic effects on normal breast epithelial cells resulted in a 50% reduction in new contralateral breast cancers, evidence ligands, including estrogen and tamoxifen, in different tissues that provided much of the impetus to developing a role for (reviewed in Refs. 5, 6). Estrogen-regulated gene transcription is tamoxifen in chemoprevention. At the same time, the estrogenic activated through two separate transactivation domains within effects of tamoxifen therapy on bone and cholesterol are of ER, termed AF-1 in the NH2-terminal A/B region and AF-2 in clinical benefit in reducing risk from osteoporosis and cardio- the COOH-termihal E region (18). At its simplest level, tamox- vascular disease (11, 12). In the adjuvant setting, tamoxifen's ifen functions as a competitive antiestrogen to inhibit estrogen increased risk of endometrial cancer has been perceived as small action. Tamoxifen-bound ER still dimerizes and binds DNA, in relation to the substantial benefit from reduction in breast although the downstream effects differ because of the altered cancer-related events (14). However, both in adjuvant and met- conformational shape of the tamoxifen-ER complex compared astatic therapy with tamoxifen, breast epithelial cells and estab- with estradiol, which induces a change in the receptor-bound lished tumors adapt to chronic antiestrogen exposure and de- balance of coactivators and corepressors, thus blocking gene velop resistance to tamoxifen, which may relate to the partial transcription through the AF-2 domain while still allowing AF-1 agonist effect of tamoxifen in stimulating tumor growth (re- mediated gene transcription to occur (19). This may explain the viewed in Refs. 15, 16). Experimental models have shown that partial agonist activity of tamoxifen in addition to its ability to novel antiestrogens that are devoid of agonist effects can antag- antagonize estrogen-regulated gene transcription (Fig. 1), onize tamoxifen-stimulated growth (9) and, as treatment of It has become clear that the molecular biology of ER is hormone-sensitive tumors, may delay the emergence of resist- complex, and that other aspects of its function may mediate the ance (17). These observations generated hope that better SERMs differential ligand effects seen in response to estrogen or tamox- with an improved antiestrogen/estrogen profile may overcome ifen. In addition to classical ER (now called ERa), a second ER this form of resistance and improve further on the efficacy of was cloned (ER[3), which shares sequence homology within the tamoxifen in treating breast cancer. DNA-binding domain (20) but which differs in that AF-1 activity is considerably less than with ERa (21). Equally, ER[3 lacks much of the COOH-terminal F domain of ERa, which ER Molecular Biology: Differential Effects may be an important region in determining an agonist response of SERMs to tamoxifen (22). Recent evidence has implicated increased Progress in our molecular understanding of ER function ER[3 expression as a mechanism for tamoxifen resistance in has provided insights into the differential effects of various ER breast cancer (23). Other response pathways may become acti-

1.3

ci I OH

Tamoxifen Toremifene Droloxifene Idoxifene Fig. 2 Structure of tamoxifen- like SERMs.

O~H

? HO-P-O

TAT-59 GW5638 Lasofoxifene

vated indirectly by ER, including AP-I response elements, scriptional response to liganded ER for a given gene (31). This which regulate genes involved in cell proliferation, motility, and provides a new hypothesis based on the chemical structure of apoptosis. Tamoxifen was shown to be an agonist on AP-1- each of the different SERMs that may explain their differential regulated genes with either ERoL or ER[3 (24), whereas estrogen antagonist/agonist profiles. liganded with ER[3 inhibited AP-1 gene transcription (25). En- hanced activation of AP-1 by tamoxifen may also be associated with tamoxifen resistance in models of breast cancer (26) and in Novel SERMs: Potential Advantages for tumors from breast cancer patients relapsing on tamoxifen (27). Breast Cancer Finally, the relative balance in a given cell type of coactivator An understanding of how the triphenylethylene antiestro- and corepressor proteins may also determine the response of ER gen tamoxifen interacts with ER has allowed novel SERMs to to a particular ligand (28), and a reduction in the level of the be synthesized that possess an improved antiestrogenic/estro- corepressor N-CoR has been associated with the development of genic profile. These drugs have been developed with the aim of tamoxifen resistance in breast cancer xenografts (29). retaining both the antagonist activity of tamoxifen within the The development of SERMs that are structurally different breast and the agonist profile in bone and the cardiovascular from tamoxifen has provided an additional insight into the system, yet at the same time eliminating unwanted agonist biology of ER action (4). A crucial aspect of estrogen interaction effects on the gynecological tract, in particular the uterus. Non- with ER is the complete envelopment of the steroid in a hydro- steroidal SERMs fall into two broad categories: those that are phobic pocket within the ligand-binding domain because of the structurally similar to the triphenylethylene structure of tamox- critical positioning of a "lid" formed by helix 12 of the ER. The ifen (first-generation SERMs; Fig. 2), and those that are struc- position of helix 12 is also critical for recruitment of coactiva- turally different and more related to the benzothiophene struc- tors to the AF-2 transactivation site, allowing subsequent initi- ture of raloxifene (second- and third-generation SERMs; Fig. 3). ation of RNA polymerase activity. SERMs or steroidal anties- A third class of antiestrogen includes the steroidal antiestrogen trogens may result in a qualitatively different conformational ICI-182,780 (fulvestrant), which is a structural derivative of shape of the liganded receptor, because of the alkylaminoethoxy estradiol with a long hydrophobic side chain at the 7o~ position side chain of the triphenylethylenes (e.g., tamoxifen), the dif- (Fig. 3; Ref. 32). Pharmacologically, these latter compounds are ferent structure of the benzothiophenes (e.g., raloxifene), or the pure antiestrogens that not only impair ER dimerization but also long side chain of steroidal antiestrogens (fulvestrant; Fig. 1). induce ER degradation (33, 34) and, thus, act as potent anties- This imparts a different positioning of the helix 12 lid, the exact trogens in all tissues including the breast, uterus, and probably nature of which depends on the conformational shape each bone. Although some may argue that fulvestrant is not a true antiestrogen imparts to the ligand ER complex (30). As a result, SERM because it lacks selective agonist/antagonist effects in the relative positioning of helix 12 may influence the likelihood different tissues and possesses a fundamentally different mech- of coactivator/corepressor binding and, thus, determine the tran- anism of action, others have suggested that it represents one

0 0 0

Fig. 3 Structure of second Raloxifene and third-generation SERMs. LY 353381 (SERM 3) ERA-923 (WAY-138923)

Raloxifene (redrawn) EM 652 (SCH 57068) Fuivestrant (ICI 182,780)

extreme (i.e., a pure antiestrogen) of the SERM spectrum, with Table 1 The ideal profile of a novel SERM in comparison estrogen as a pure agonist at the other end, and all other SERMs with tamoxifen falling somewhere in between (5). Preclinical Each of the SERMs demonstrated pharmacological or Greater binding affinity for ER Ability to antagonize estrogen-dependent growth of breast cancer pharmacodynamic benefit over tamoxifen in various preclinical cells in vitro studies, and as a consequence had a profile that supported Equal or greater inhibition of hormone-dependent xenograft growth clinical development in women with advanced breast cancer in in vivo the hope of producing a more effective and beneficial antiestro- Activity against tamoxifen-stimulated (resistant) tumors gen. The potential preclinical advantage for these SERMs in- Delayed emergence of antiestrogen resistance in vivo Reduced agonist effects in uterotrophic assays cluded either greater potency attributable to enhanced affinity Lack of stimulation of endometrial cancer cells in vitro/in vivo for ER, greater efficacy compared with tamoxifen against breast Lack of DNA adduct formation cancer in vitro or in vivo, or reduced risk of toxicity compared Prevention of bone loss in ovariectomized animals with tamoxifen on organs such as the liver and endometrium Clinical Activity in hormone-sensitive breast cancer, at least equivalent to (Table 1). If resistance to tamoxifen occurs in part because of tamoxifen the agonist effects of the drug that stimulate tumor regrowth (15, Increase in time-to-disease progression compared with tamoxifen 16), then SERMs would be expected either to be active against Activity in tamoxifen-resistant breast cancer tamoxifen-resistant tumors or to delay the emergence of resist- Improved side effect profile (e.g., reduced hot flushes) ance. In the clinic, this profile would manifest as either a No endometrial thickening/hyperplasia/cancer risk Preservation of bone mineral density superior response rate or a delay in the emergence of resistance Reduction in serum cholesterol during long-term therapy. Thus, one might expect to see evidence of activity for SERMs in Phase II studies in tamoxifen- resistant breast cancer or, alternatively, an increased duration of clinical response or time-to-disease progression compared profile compared with tamoxifen led to their clinical develop- with tamoxifen in randomized Phase III trials as first- ment with the hope that these agents might prove safer or more line therapy for ER+ve hormone:sensitive breast cancer effective antiestrogens for the treatment of breast cancer. (Table 1). The progress to date with each SERM compound Torernifene. Toremifene's only structural difference is reviewed below, in particular recent data from clinical compared with tamoxifen relates to a single chlorine atom at trials of SERMs in women with either tamoxifen-resistant or position 4 (Fig. 2), and the pharmacological profiles of both hormone-sensitive breast cancer. drugs are similar (35, 36). Unlike tamoxifen, toremifene was found not to be hepatocarcinogenic in preclinical models (37, Tamoxifen-like SERMs 38), which in part may relate to an inability of toremifene Of the triphenylethylene derivatives, clinical data from compared with tamoxifen to induce DNA adducts in the rat liver Phase IUIII clinical trials in women with advanced breast cancer (39). Toremifene had a similar relative binding affinity for ER have been published with three triphenylethylene tamoxifen-like compared with tamoxifen and inhibited the growth of ER+ve compounds (toremifene, droloxifene, and idoxifene). For each breast cancer cells in vitro (40) and hormone-dependent breast one, preclinical data that had suggested an improved SERM cancer xenograft growth in vivo (41). However, like tamoxifen,

Table 2 Summary of clinical efficacy data from the randomized Phase III trials of toremifene (40-60 mg/day) versus tamoxifen (20-40 mg/ day) as first-line endocrine treatment of advanced breast cancer in postmenopausal women (ER status positive or unknown) Toremifene Tamoxifen Study n ORRa TTP (mo) n ORR (%) TTP (mo) Hayes et al. (53) 221 21% 5.6 215 19% 5.8 Pyrhonen et aL (54) 214 31% 7.3 201 37% 10.2 Gershanovich et al. (55) 157 21% 4.9 149 21% 5.0 Nomura et al. (52) 62 24% 5.1 60 27% 5.1 Milla-Santos et al. (56) 106 38% 11.9 111 32% 9.2 Meta-analysis b (57) 725 24.0% 4.9 696 25.3% 5.3 "ORR, objective response rate, including complete response and partial response; TTP, median time-to-disease progression. b The meta-analysis (57) was published in 1999 and included data from the first four trials, together with a small German study (M. Kaufman, unpublished data), but did not include the Spanish study (56), which was published only in 2001.

toremifene had estrogenic effects on both endometrial cells and Any potential difference in carcinogenicity identified in the uterus in vivo (42, 43), although it had slightly reduced preclinical studies was not evaluated in any of these studies and estrogenic effects on bone (44). Toremifene was developed, is probably of relatively little clinical significance in advanced therefore, as a triphenylethylene derivative of tamoxifen that breast cancer. However, at least two adjuvant studies were might have less genotoxic potential and, therefore, could be a subsequently initiated to compare efficacy and, in particular, safer antiestrogen for breast cancer treatment. long-term tolerability and safety in early breast cancer patients. High-dose toremifene (120-240 mg) has been investigated Preliminary data from --900 postmenopausal node-positive pa- in five Phase II studies as second-line therapy in a total of 260 tients after a median follow-up of 3.4 years have been reported, patients with tamoxifen-resistant breast cancer (45-49). These and there were no significant differences in efficacy or tolera- patients had failed to respond to tamoxifen for advanced disease, bility compared with tamoxifen (58). In particular, the number had progressed after an initial response, or had relapsed on of subsequent second cancers was similar, although longer fol- adjuvant tamoxifen. The objective response rate in these studies low-up will be needed to see if any differences emerge. ranged between 0 and 14%, although a proportion of patients Droloxifene. Structurally droloxifene is 3-hydroxyta- (17-23%) had stable disease for at least 6 months. Although moxifen and, as such, has a 10-fold higher relative binding occasional tamoxifen-refractory patients may have had an ob- affinity for ER compared with tamoxifen (59). In preclinical jective response to toremifene, especially if they had responded studies, droloxifene had several potential advantages over ta- to tamoxifen previously, it was concluded that cross-resistance moxifen, including a shorter half-life (60), greater growth inhi- probably exists between the two drugs (50). bition of breast cancer cells and reduced estrogenicity in the rat With toremifene as first-line therapy in hormone-sensitive uterus (61), and absence of DNA adduct formation or carcino- advanced breast cancer, several Phase II studies in a total of 175 genicity (62). However, like tamoxifen, it also behaved as an patients showed objective responses rates of 48-68% with estrogen in bone, preserving bone mineral density (63). doses of 60-240 mg daily, with a suggestion that higher re- Early Phase HI studies suggested some efficacy in patients sponse rates occurred with the 240-mg dose (reviewed in Ref. who had received tamoxifen previously (64, 65). A Phase II 50). Low-dose (20 mg) toremifene was associated with a re- study of droloxifene, 100 mg daily, in 26 patients who had sponse rate of only 21% in an additional small study and was not received tamoxifen previously, found a response rate of 15%, investigated further (51). Subsequently, there have been five with stable disease >6 months in an additional five (19%) large Phase III randomized controlled trials of toremifene patients (66). A large randomized dose-finding study of 20, 40, (40-60 mg) versus tamoxifen (20-40 mg) as first-line endo- or 100 mg droloxifene as first-line therapy in 369 patients crine therapy in advanced breast cancer (see Table 2; Refs. showed objective response rates of 30, 47, and 44%, respec- 52-56). The response rate to toremifene in these larger multi- tively (67, 68). Better response duration and time-to-disease center studies was lower than in the Phase II studies and ranged progression were seen with the two higher doses, and there were from 21 to 38%. In all of these studies, toremifene showed no significant drug-related toxicities. Other first-line Phase II efficacy equivalent to tamoxifen for objective response rate, studies were undertaken, including one study in 39 patients that stable disease, time-to-disease progression, and overall survival showed a response rate of 51% (95% CI, 35-67%) and median (Table 2). In addition, two of these studies randomized patients time-to-progression of 8 months (69). These first-line data sug- between 60 mg and higher doses (200/240 mg) of toremifene gested a level of efficacy comparable with that expected with and found no significant difference in efficacy (53, 55). There tamoxifen, and randomized Phase III studies comparing drolox- was no difference in drug-related toxicities, and both toremifene ifene versus tamoxifen were initiated. However, droloxifene and tamoxifen were well tolerated. A recent meta-analysis of was found to be less active than tamoxifen, and further devel- 1421 patients from these trials showed a similar response rate opment was stopped (70, 71). for toremifene compared with tamoxifen (24 versus 25.3%), ldoxifene. Idoxifene (Fig. 2) is a SERM that is metabol- with no significant difference in time-to-disease progression ically more stable than tamoxifen as a result of a pyrrolidino (hazard ratio, 0.98; 95% CI, 0.87-1.11) or overall survival side chain, with increased binding affinity for ER because of the (hazard ratio, 0.98; 95% CI, 0.83-1.15; Ref. 57). substitution of an iodine atom at the 4 position (72). Idoxifene

Table 3 Overall efficacy of tamoxifen-like SERMs in advanced breast cancer: response rate ranges from Phase II studies of toremifene, droloxifene, and idoxifene in tamoxifen-resistant or hormone-sensitive patients, and Phase III trials in first-line versus tamoxifen Tamoxifen-resistant, Phase II Hormone-sensitive Phase II Phase III Phase II/III ORW' (%) S.D. (%) ORR (%) ORR (%) TTP (%) Toremifene 0-14 16-30 21-68 21-38 4.9-11 Droloxifene 15 19 30-51 Idoxifene 9 9 20 6.5 Median 5 18 31 6.9 '70RR, % patients with an objective response, including complete response and partial response; S.D., % patients with stable disease for -----6 months; TTP, median time to disease progression (in months).

inhibited hormone-dependent breast cancer growth and was that demonstrated significantly reduced agonist activity on the more effective than tamoxifen at inhibiting both MCF-7 cell uterus in ovariectomized rats, yet remained a full agonist in growth in vitro and rat mammary tumor growth in vivo (73). As reducing cholesterol and maintaining bone mineral density (81). a SERM, idoxifene had estrogenic agonist effects on bone (74). CGP 336,156 (lasofoxifene) is a derivative of tetrahydronaph- However, reduced agonist activity on breast cancer cells for thalene that maintains bone mineral density in animal models idoxifene compared with tamoxifen in vivo was suggested by (82) and, thus, may find an application for the prevention of the greater inhibition of a MCF-7 xenograft growth in the osteoporosis (6). There are few (if any) published clinical data absence of estradiol (75). Likewise, reduced stimulation of for any of these compounds in advanced breast cancer. uterine weight was seen in various uterotrophic assays (73, 74). Clinical Efficacy of Tamoxifen-like SERMs. From the Thus, idoxifene was developed in the hope that the reduced clinical data following failure of tamoxifen in advanced breast agonist profile in breast and gynecological tissues would pro- cancer, little significant activity has been observed overall with vide an advantage over tamoxifen for breast cancer patients. the first-generation tamoxifen-like SERMs (toremifene, drolox- In a Phase I study of idoxifene in 14 patients who had irene, and idoxifene) with a median response rate from all previously received tamoxifen, 2 patients had a partial response studies of only 5% (range, 0-15%; Table 3). The reduced with idoxifene, and 3 patients had disease stabilization for >6 agonist profile seen with droloxifene and idoxifene in preclini- months (76). Results from a randomized Phase II study showed cal studies may have been tissue- or cell-specific and did not little evidence of significant clinical activity for idoxifene in 56 appear to manifest itself as any improved efficacy in treating or postmenopausal patients with tamoxifen-resistant breast cancer preventing tamoxifen resistance in patients with breast cancer. If (77). Two partial responses (objective response rate, 9%) and the agonist activity of tamoxifen were a major mechanism for two patients with stable disease were seen with idoxifene, the development of resistance, one might have hoped that whereas, in contrast, no objective responses were seen with a SERMs with reduced agonist activity would offer a longer higher (40-mg) dose of tamoxifen. In a Phase III trial a total of response duration or time to progression. The fact that they did 220 postmenopausal women with metastatic breast cancer were not implies that, unlike the steroidal antiestrogen ICI 182,780 randomized to receive either idoxifene 40 mg/day or tamoxifen (fulvestrant), these drugs are probably completely cross-resis- 20 mg/day as first-line endocrine therapy (78). The objective tant with tamoxifen. Perhaps this is not surprising given the response rate [complete response plus partial response similar tamoxifen-like mechanism of action and structure-func- (CR+PR)] was 20% (95% CI, 12.7-28.2%) for idoxifene and tion interaction with ER for these triphenylethylene compounds. 19% (95% CI, 12.5-28.2%) for tamoxifen, with a median du- In contrast, fulvestrant acts by down-regulating ER expression ration of objective response of 8.1 months for idoxifene and 7.3 (33, 34), and this may explain why the drug appears to have months for tamoxifen. There was no significant difference in much better activity in tamoxifen-resistant breast cancer than time-to-disease-progression or overall survival. Although no toremifene or idoxifene (83, 84). differences were reported in gynecological adverse events be- The combined Phase II/III clinical trial data for tamoxifen- tween idoxifene and tamoxifen, in a parallel osteoporosis pro- like SERMs (toremifene, droloxifene, idoxifene) as first-line gram an increased incidence of uterine prolapse and polyps was therapy suggest a median response rate of 31% (range, 20- reported in idoxifene-treated women. Thus, despite a reduced 68%), with a median time-to-disease progression of 6.9 months agonist profile for idoxifene seen in preclinical studies, there (Table 3). In the randomized first-line trials in hormone-sensi- appeared to be no major differences in terms of clinical efficacy tive advanced breast cancer, both toremifene and idoxifene were or safety profile between idoxifene and tamoxifen, and further shown to be very similar to tamoxifen in terms of both clinical development of the drug was stopped. efficacy and toxicity (52-56, 78), whereas droloxifene appeared Other Tamoxifen-like Derivatives in Development. to be inferior (70). The toxicity profile was the same, including Other structural analogues of tamoxifen (Fig. 2) have been gynecological effects seen with idoxifene. On the basis of these synthesized, including TAT-59, which has a 10-fold higher present data, therefore, it is unlikely that the first-generation affinity for ER than tamoxifen and was more effective at inhib- triphenylethylene SERMs will replace tamoxifen for advanced iting human breast cancer xenograft growth in vivo (79, 80). breast cancer, because they have failed to show superiority or GW5638 is a carboxylic derivative in early clinical development any significant clinical advantage.

Novel Second- and Third-Generation SERMs arzoxifene has entered clinical development for the treatment of Greater optimism has surrounded the profile of second- and breast cancer. third-generation SERMs (Fig. 3), in particular, that this profile In a Phase I study, 32 patients who had received a median may translate into an improved clinical benefit for breast cancer of two prior endocrine therapies were treated with arzoxifene in patients. Much of the enthusiasm relates to the fact that these doses ranging from 10 to 100 mg daily (99). No significant drugs appear devoid of any agonist activity in the endometrium, toxicities were seen, and, in particular, transvaginal ultrasound while at the same time appearing to be potent antiestrogens in showed no endometrial thickening after 3 months of therapy. the breast that retain agonist activity in bone. Structurally, most Six patients had stable disease for a median of 7.7 months (range, 6-33 months). In a Phase II study with arzoxifene as of these drugs resemble the benzothiophene raloxifene which is first-line therapy, 92 patients were randomized to either 20 mg the most extensively studied SERM in this class. or 50 mg daily (100). Only 95 patients had received tamoxifen Raloxifene. The binding affinity of raloxifene for ER is previously in the adjuvant setting. There was no difference in similar to that of tamoxifen (85), and most of the pharmacolog- response rate (36 versus 34%), in clinical benefit rate that ical data showed activity that was similar to tamoxifen in terms included stable disease (63 versus 64%), or in time-to-disease of inhibiting breast cancer cells in vitro and rat mammary tumor progression (10.4 versus 8.9 months). Likewise toxicities were growth in vivo (86, 87). In preclinical models, the drug main- minor, although 30% reported minor hot flashes. More recently, tained bone mineral density (8) and reduced total cholesterol preliminary results were reported of a further Phase II trial that (88) but, compared with tamoxifen, had significantly less estro- compared both doses (20 and 50 mg) in 63 tamoxifen-resistant genic activity on endometrial cells and could inhibit tamoxifen- patients, and separately in 49 patients with hormone-sensitive stimulated endometrial cancer growth in vivo (9). Raloxifene disease (i.e., first-line therapy; Ref. 101). Response rates were was subsequently developed and is now indicated for osteopo- low in the tamoxifen-resistant patients (10% for 20 mg, 3% for rosis based on clinical trials that showed prevention of bone loss 50 mg), all of whom either had relapsed on adjuvant tamoxifen in postmenopausal women (89). after at least 1 year of therapy or had progressed on tamoxifen Although raloxifene was not developed as an antiestrogen for advanced disease after an initial response. In contrast, a for breast cancer, some limited data exist on the activity of response rate of 30% was seen with 20-mg arzoxifene in the raloxifene in patients with advanced breast cancer. In a small hormone-sensitive group, with an additional 17% having stable study in 14 patients who had become resistant to tamoxifen after disease and an overall median time-to-progression of 8.3 an initial response, only 1 patient had a minor response when months. The response rate for the 50-mg dose was somewhat treated with 200 mg of raloxifene (90). In 21 patients with lower (8%), although numbers were small (only 25 patients). On ER+ ve metastatic breast cancer treated with raloxifene 150 mg the basis of all of the Phase II data, 20-mg arzoxifene has now twice a day as first-line therapy, 4 (19%) had a partial response been taken forward into a large multicenter Phase III trial for a median duration of 22 months with an additional 3 (14%) against tamoxifen as first-line therapy. patients showing stable disease (91). Raloxifene does not appear EM-800. This is an orally active so-called pure nonste- to relieve vasomotor symptoms such as hot flashes. However, roidal antiestrogen that is a prodrug of the active benzopyrene during raloxifene's development for osteoporosis, it was found derivative EM-652 (SCH 57068; Ref. 102). The binding affinity to significantly reduce the incidence of breast cancer (in partic- of EM-652 for ER is significantly greater than estradiol, tamox- ular, ER+ve tumors) in postmenopausal women by 76% (95% ifen, raloxifene, or fulvestrant (103). The prodrug EM-800 is a CI, 56-87%), without any increase in endometrial thickening or potent antiestrogen and was more effective than 4-hydroxyta- risk to the gynecological tract (92, 93). Because tamoxifen may moxifen and fulvestrant at inhibiting estradiol (E2)-induced cell also reduce breast cancer incidence, albeit with an increased risk proliferation in breast cancer cells in vitro, and, in the absence of endometrial cancer and thrombotic events (3), the current of E 2, had no agonist effects on growth (104). In ZR-75-1 Study of Tamoxifen and Raloxifene (STAR) chemoprevention xenografts, EM-800 was significantly more effective than ta- trial is comparing the effects of raloxifene with tamoxifen. The moxifen at inducing tumor regressions in vivo and, in the ab- potential exists that, as a SERM, raloxifene may reduce breast sence of Ex, antagonized tamoxifen-stimulated tumor growth cancer incidence with a better safety profile compared with (105). In intact mice, EM-800 was 30-fold more potent than tamoxifen; it is hoped that this trial will clarify which patients tamoxifen at inhibiting uterine weight and reducing uterine/ (i.e., with what level of breast cancer risk) derive benefit from vaginal ER expression (106). Likewise, EM-800 was devoid of chemoprevention. any stimulatory effect on alkaline phosphatase activity (a sen- Arzoxifene. LY 353381 (arzoxifene) is a benzothio- sitive marker of estrogenic activity) in Ishikawa endometrial phene analogue, which is a more potent antiestrogen with an carcinoma cells (107), whereas EM-652 had no agonist activity improved SERM profile compared with raloxifene (94). In in an immature rat uterotrophic assay (108). In addition, studies particular, arzoxifene was a more potent inhibitor of breast have shown that EM-800 prevented bone loss in the ovariecto- cancer cells in vitro than either tamoxifen or raloxifene and mized rat (109) and lowered serum cholesterol levels (102). inhibited the growth of mammary tumor xenografts in vivo (95, Interestingly, EM-800 appears to significantly down-regulate 96). As a SERM, in preclinical studies, arzoxifene was a more ER levels both in tumors and normal estrogen-sensitive tissues potent agonist on bone and cholesterol metabolism than ralox- in a similar fashion to the steroidal antiestrogen fulvestrant ifene (97, 98), with no evidence of any estrogen-like agonist (106), but its specific agonist effects on bone differentiate it effects on uterine tissues (94). In view of these promising data, from fulvestrant, which has not been shown to prevent bone

Table 4 Overallefficacy of second- and third-generation SERMs in IUIII trials with arzoxifene and ERA-923 are awaited, but to advanced breast cancer: response rate ranges from clinical trial results date, there is little clinical evidence to suggest that in advanced from Phase II studies of raloxifene, arzoxifene, and EM-800 in breast cancer, substantial improvements in efficacy will be made tamoxifen-resistant or hormone-sensitive patients over tamoxifen. Tamoxifen-resistant Hormone-sensitive Phase II Phase II ORRa (%) S.D. (%) ORR (%) TTP (%) Conclusion: Future Role for SERMs in Raloxifene 0 19 Breast Cancer Arzoxifene 3-10 3-7 30-36 8.3-10.4 It is unclear to what extent any preclinical advantages that EM-800 14 23 have been observed for each of these SERMs over tamoxifen Median 6.5 7 30 9.4 may be predictive of clinical outcome in the treatment of ad- "ORR, % patients with an objective response, including complete vanced breast cancer. Thus far, the clinical data in advanced response and partial response; S.D., % patients with stable disease for breast cancer summarized above are somewhat disappointing ->6 months; TTP, median time to disease progression (in months). for the first-generation tamoxifen-like SERMs. Instead, much greater potential may exist in either the adjuvant or chemopre- ventive setting in which an improved SERM profile on bone, lipid metabolism, and the endometrium will be of maximum loss. As such, EM-800/EM-652 has a potentially promising benefit. It remains to be seen whether vasomotor symptoms SERM profile. associated with both tamoxifen and raloxifene are any less In terms of clinical development, a Phase II study of frequent with the new SERMs. The dilemma faced by those EM-800 (20 mg or 40 rag) was undertaken in 43 postmeno- developing these therapies, however, is the need to demonstrate pausal women for whom tamoxifen had failed either in the clinical activity against breast cancer that is at least equivalent to metastatic or adjuvant setting (102, 110). There was one com- tamoxifen. The clinical data outlined above suggest that, al- plete response and five partial responses (response rate, 14%), though there is probably little role for these drugs after failure of with most of the responses occurring in those who had received tamoxifen, their efficacy and tolerability in hormone-sensitive at least 3 years adjuvant tamoxifen (110). An additional 10 advanced breast cancer is probably equivalent to that of tamox- (23%) patients had stable disease for >6 months. On the basis ifen. Are these sufficient data to justify expensive large-scale of these results, a randomized Phase III study in patients who adjuvant programs? had failed to respond to tamoxifen was undertaken, comparing To complicate matters further, the scenario in hormone- the efficacy of EM-800 (20 mg or 40 mg) with the third- dependent advanced breast cancer has changed significantly generation aromatase inhibitor anastrozole. At the defined in- within the last 6-12 months. It is now clear that third-generation terim review when over 300 patients had been entered, the aromatase inhibitors (e.g., letrozole, anastrozole) are probably efficacy was less than that of anastrozole, and the trial was superior and better tolerated than tamoxifen (112-114). As a terminated. There are no data at present on the activity of consequence, it is now difficult to envisage whether SERMs will EM-800 in the first-line hormone-sensitive population. have any role as first-line therapy for postmenopausal women ERA-923. ERA-923 is a novel SERM that appears to with advanced breast cancer. If aromatase inhibitors become the have an improved preclinical profile compared with tamoxifen first-line postmenopausal endocrine therapy of choice in the and raloxifene (111). ERA-923 is now being evaluated in a adjuvant as well as the metastatic setting, the question will soon randomized dose-finding Phase II trial (25 versus 100 mg) as emerge if or when antiestrogens such as tamoxifen, SERMs, or second-line therapy in 100 ER+ve patients with tamoxifen- the steroidal compound fulvestrant should be used in breast resistant metastatic breast cancer. A similar randomized Phase II cancer treatment. In vitro it is known that breast cancer cells trial has been proposed in receptor-positive hormone-sensitive adapt when subjected to long-term estrogen deprivation, remain- metastatic breast cancer as first-line therapy. ing ER+ve and becoming hypersensitive to very low concen- Clinical Efficacy of Second- and Third-Generation trations of E 2 (115). It is conceivable that potent antiestrogens SERMs. These new compounds in preclinical models appear including SERMs could be active in this setting, and clinical to offer a greater increase in potency and tumor growth inhibi- trials with fulvestrant after aromatase inhibitor failure are in tion, together with an improved SERM profile on other tissues, progress. in comparison with the tamoxifen-like SERMs. At the present An alternative role for SERMs could be as adjuvant ther- time, there are too few clinical data to know whether these apy, either alone or in combination with aromatase inhibitors, potential advantages will translate into beneficial effects for thus providing protection to the bone and cardiovascular system breast cancer patients. However, in tamoxifen-resistant patients, while enhancing antitumor efficacy. However, it is not clear the reported level of activity is low for raloxifene (90), arzox- whether any potential negative interaction on tumor cell growth ifene (101), and EM-800 (Ref. 102; Table 4), with a median could occur because of agonist activity of SERMs in an envi- response rate of 6.5%, which is very similar to that observed ronment of complete estrogen deprivation. It has always been with the tamoxifen-like SERMs (Table 3). It is probable that thought that endocrine therapies are better given in sequence activity in first-line therapy will be similar to that of tamoxifen, rather than in combination, a view that has been challenged because the only Phase II data with raloxifene and arzoxifene recently by data in premenopausal ER+ve advanced breast give a median response rate of 30%, with a median time to cancer in which combined estrogen deprivation with an anties- progression of 9.4 months (Table 4). Results of ongoing Phase trogen was superior to either therapy alone, including a benefit

in overall survival (116). To develop SERMs as adjuvant ther- both scenarios, so we just have to wait and see what the STAR apy, including as combined therapy with aromatase inhibitors, study shows. there seems to be no shortcut to performing some form of Dr. Anthony Howell: That's right. But the women treated clinical efficacy/safety study in advanced breast cancer. Addi- in the MORE study were all old, whereas in the P1 study they tional evidence of a SERM's biological activity and clinical ranged in age, and it may be that these agents are better in older efficacy could be ascertained from short-term randomized neo- people, and the difference is a serious difference. adjuvant studies, as undertaken for idoxifene (117), raloxifene Dr. Victor Vogel: Just a comment on the MORE study. (118), and fulvestrant (119). The next 5 years will be crucial to Not only were those patients old, but they were old with osteo- see whether the latest generation of SERMs have a significant porosis. There's a summary of osteoporosis data in the current role in breast cancer therapy, and more importantly, what that issue of the Journal of the National Cancer Institute showing role might be. that with demonstrable osteoporosis, the breast cancer risk is about 3-fold lower [J. M. Zmuda et al., J. Natl. Cancer Inst. (Bethesda), 93: 930-936, 2001]. Now, whether or not that Open Discussion would translate into an inability of raloxifene to reduce risk is Dr. Stephen Johnston: Droloxifene went into randomized exactly the question in the STAR trial. studies against tamoxifen, but its development was stopped. I Dr. Paul Goss: It's true osteoporosis is a reduced risk don't know if anyone has any information on that. factor, but when you actually look at the incidence of breast Dr. Aman Buzdar: There were more than 1100 patients cancer risk in the placebo arm of the MORE trial, it's higher randomized to tamoxifen versus droloxifene, and there was no than the SEER population-based age-matched incidence. The time-to-progression advantage in favor of the new compound. trial still attracted women with an increased risk of breast So the study was stopped, and the manuscript was submitted. cancer, despite early osteoporosis. The first journal thought it was a very good study, but because Dr. Matthew Ellis: Dr. Brodie has a very nice aromatase it's a negative study, they didn't want to publish it. inhibitor model of the transfected MCF-7 cells. What about in Dr. Kathleen Pritchard: Here's an example of perfectly the tamoxifen-resistant tumor models in nude mice? That model good data that should come out not getting published anywhere. also is not quite as robust, although at least it predicted fulves- I wondered, when I looked at the data from that study, whether trant worked in tamoxifen-resistant breast cancer. it was another example of choosing a dose that may not have Dr. Osborne: I think tamoxifen-stimulated growth is a been high enough. Because what else is the explanation? You common source of resistance, and I think it predicts what have a very big study with a lot of patients, and you see this happens with different endocrine therapies. phenomenon where the drug worked poorly in the younger Dr. Johnston: You have to be careful with that model; we group of patients where maybe there was competing estrogen, did some studies with idoxifene in that scenario as well. We whereas in the patients over 60, the drugs were essentially were able to show partial non-cross-resistance. Some of the equivalent. We had a small dose-finding study, and the two tumors that grew with tamoxifen were more likely to respond to higher doses looked the same. One wasn't more toxic than the idoxifene. But it wasn't dramatic by any means-not as convinc- other, and we decided to go with the lower dose because it ing as the fulvestrant data. It's a model. But when we've gone looked just as good. to look in patients, we just don't see the response. Dr. Per Lonning: It raises the question, how do we actually do a dose selection for the SERMs? All of the surrogate parameters we know are related to the estrogen agonistic, not the References antagonistic effects. 1. Cole, M. P., Jones, C. T. A., Todd, I. D. H., et al. A new antiestro- Dr. Johnston: The other way to do that is to do the genic agent for breast cancer. An early appraisal of ICI 46,474. Br. J. short-term preoperative study. You can look at proliferative or Cancer, 25: 270-275, 1971. apoptotic changes in the primary tumor. You can do it in far 2. Early Breast Cancer Trialists Group. Tamoxifen for early breast cancer: an overview of the randomised trials. Lancet, 351:1451-1467, fewer patients, and you can probably get a cleaner study. But 1998. then it's an issue of whether you can take a brand new SERM 3. Fisher, B., Costantini, J. P., Wickerham, D. L., et al. Tamoxifen for and put it into earlier stage patients, because you need safety or the prevention of breast cancer: report of the National Surgical Adjuvant toxicity data prior to that. Breast and Bowel Project P-1 study. J. Natl. Cancer Inst. (Bethesda), 90: 1371-1388, 1998. Dr. Kent Osborne: Are you looking at the very minimal 4. Jordan, V. C. The development of tamoxifen for breast cancer data on raloxifene and breast cancer? If anything, it would seem therapy. In: V. C. Jordan (ed.), Long-Term Tamoxifen Treatment for to be a worse drug than tamoxifen, and certainly cross-resistant. Breast Cancer, pp. 3-26. Madison: University of Wisconsin Press, 1994. How do you explain the fact that the same dose seems to be so 5. Osborne, C. K., Zhao, H., and Fuqua, S. A. W. Selective estrogen effective in breast cancer prevention, or is it? receptor modulators: structure, function, and clinical use. J. Clin. On- Dr. Johnston: The first question is: is it? The second is col., 18: 3172-3186, 2000. whether or not we're looking at something completely different 6. Levenson, A. S., and Jordan, V. C. Selective oestrogen receptor modulation: molecular pharmacology for the new millennium. Eur. J. in terms of the effect on primed or unprimed breast epithelium Cancer, 35: 1628-1639, 1999. that may or may not be malignant or premalignant versus an 7. Osborne, C. K., Hobbs, K., Clark, G. M., et al. Effects of estrogens established tumor that may be heterogeneous in terms of its and antiestrogens on growth of human breast cancer cells in athymic responsiveness. Having said that, tamoxifen is pretty effective in nude mice. Cancer Res., 45: 584-590, 1985.

8. Jordan, V. C., and Robinson, S. P. Species specific pharmacology of antiestrogen 4-hydroxytamoxifen. Mol. Endocrinol., 11: 657-666, antiestrogens: role of metabolism. Fed. Proc., 46: 1870-1874, 1987. 1997. 9. Gottardis, M. M., Ricchio, M. D., Satyaswaroop, P. G., et al. Effect 29. Lavinsky, R. M., Jepsen, K., Heinzel, T., et al. Diverse signaling of steroidal and nonsteroidal antiestrogens on the growth of a tamoxifen- pathways modulate nuclear receptor recruitment of N-CoR and SMRT stimulated human endometrial carcinoma (EnCal01) in athymic mice. complexes. Proc. Natl. Aead. Sci. USA, 95: 2920-2925, 1998. Cancer Res., 50:3189-3192, 1990. 30. Brzozowski, A. M., Pike, A. C. W., Dauter, Z., et al. Molecular 10. Love, R. R., Cameron, L., Connell, B. L., et al. Symptoms associ- basis of agonism and antagonism in the estrogen receptor. Nature ated with tamoxifen-treatment in postmenopausal women. Arch. Int. (Lond.), 389: 753-758, 1997. Med., 151: 1842-1847, 1991. 31. Shiau, A. K., Barstad, D., Loria, P. M., et al. The structural basis of 11. Rutquivst, L. E., and Mattsson, A. Cardiac and thromboembolic estrogen receptor/coactivator recognition and the antagonism of this morbidity among postmenopausal women with early-stage breast cancer interaction by tamoxifen. Cell, 95: 927-937, 1998. in randomized trials of adjuvant tamoxifen. J. Natl. Cancer Inst. (Be- 32. Wakeling, A. E. Similarities and distinctions in the mode of action thesda), 85: 1398-1406, 1993. of different classes of antioestrogens. Endocr.-Rel. Cancer, 7: 17-28, 12. Love, R. R., Mazess, R. B., Barden, H. S., et al. Effects of tamox- 2000. ifen on bone mineral density in postmenopausal women with breast 33. Dauvois, S., Daniellan, P. S., White, R., et al. Antiestrogen ICI cancer. N. Engl. J. Med., 326: 852-856, 1992. 164,384 reduces cellular estrogen content by increasing its turnover. 13. Powles, T. J., Hickish, T., Kanis, J. A., et al. Effect of tamoxifen on Proc. Natl. Acad. Sci. USA, 89: 4037-4041, 1992. bone mineral density measured by dual-energy x-ray absorptiometry in healthy premenopausal and postmenopausal women. J. Clin. Oncol., 14: 34. Parker, M. G. Action of pure antiestrogens in inhibiting estrogen 78-84, 1996. receptor function. Breast Cancer Res. Treat., 26: 131-137, 1993. 14. Fisher, B., Costantini, J. P., Redmond, C. K., et al. Endometrial 35. Robinson, S. P., Parker, C. J., and Jordan, V. C. Preclinical studies cancer in tamoxifen-treated breast cancer patients: findings from the with toremifene as an antitumour agent. Breast Cancer Res. Treat., 16: National Surgical Adjuvant Breast and Bowel Project (NSABP) B-t4. 9-17, 1990. J. Natl. Cancer Inst. (Bethesda), 86: 527-537, 1994. 36. DiSalle, E., Zaccheo, T., and Ornati, G. Antiestrogenic and antitu- 15. Osborne, C. K., and Fuqua, S. A. W. Mechanisms of tamoxifen mor properties of the new triphenylethylene derivative toremifene in the resistance. Breast Cancer Res. Treat., 32: 49-55, 1994. rat. J. Steroid Biochem., 36: 203-206, 1990. 16. Johnston, S. R. D. Acquired tamoxifen resistance in human breast 37. Hard, G. C., Iatropoulos, M. J., Jordan, K., et al. Major differences cancer: potential mechanisms and clinical implications. Anticancer in the hepatocarcinogenicity and DNA adduct forming ability between Drugs, 8:911-930, 1997. toremifene and tamoxifen in female CrhCD(BR) rats. Cancer Res., 53: 17. Osborne, C. K., Coronado-Heinsohn, E. B., Hilsenbeck, S. G., et al. 4534-4541, 1993. Comparison of the effects of a pure steroidal antiestrogen with those of 38. Karlsson, S., Hirsimaki, Y., Mantyla, E., et al. A two-year dietary tamoxifen in a model of human breast cancer. J. Natl. Cancer Inst. carcinogenicity study of the antiestrogen toremifene in Sprague-Dawley (Bethesda), 87: 746-750, 1995. rats. Drug Chem. Toxicol., 19: 245-266, 1996. 18. Kumar, V., Green, S., Stack, G., et al. Functional domains of the 39. White, I. N., DeMatteis, F., Davies, A., et al. Genotoxic potential of human estrogen receptor. Cell, 51: 941-951, 1987. tamoxifen and analogues in female Fischer F344/n rats, DBAJ2 and 19. Tzukerman, M. T., Esty, A., Santisomere, D., et al. Human estrogen C57BL/6 mice and in human MCL-5 cells. Carcinogenesis (Lond.), 13: receptor transcriptional capacity is determined by both cellular and 2197-2203, 1992. promoter context and mediated by two functionally distinct intramolec- 40. Kangas, L., Nieminen, A-L., Blanco, G., et al. A new triphenyleth- ular regions. Mol. Endocrinol., 8: 21-30, 1994. ylene compound, Fc-1157a. II. Antitumour effects. Cancer Chemother. 20. Kuiper, G. G., Enmark, E., Pelto-Huikko, M., et al. Cloning of a Pharmacol., 17: 109-113, 1986. novel estrogen receptor expressed in rat prostate and ovary. Proc. Natl. 41. Robinson, S. P., and Jordan, V. C. Antiestrogenic action of Acad. Sci. USA, 93: 5925-5930, 1996. toremifene on hormone-dependent, -independent, and heterogeneous 21. Cowley, S., and Parker, M. Comparison of transcriptional activation breast tumor growth in the athymic mouse. Cancer Res., 49: 1758- by ERa and ERI3. J. Steroid Biochem. Mol. Biol., 69: 165-175, 1999. 1762, 1989. 22. Montano, M., Muller, V., Trobaugh, A., et al. The carboxy-terminal 42. O'Regan, R. M., Cisneros, A., England, G. M., et al. Effects of the F domain of the human estrogen receptor: role in the transcriptional antiestrogens tamoxifen, toremifene, and ICI 182,780 on endometrial activity of the receptor and the effectiveness of antiestrogens as estrogen cancer growth. J. Natl. Cancer Inst. (Bethesda), 90: 1552-1558, 1998. antagonists. Mol. Endocrinol., 9: 814-825, 1995. 43. Tomas, E., Kauppila, A., Blanco, G., et al. Comparison between the 23. Spiers, V., Parkes, A. T., Kerin, M. J., et al. Coexpression of effects of tamoxifen and toremifene on the uterus in postmenopausal estrogen receptor a and [3: poor prognostic factors in human breast breast cancer patients. Gynecot. Oncol., 59: 241-266, 1995. cancer. Cancer Res., 59: 525-528, 1999. 44. Marttunen, M. B., Hietanen, P., Tiitinen, A., et al. Comparison of 24. Webb, P., Lopez, G. N., Uht, R. M., et al. Tamoxifen activation of the estrogen receptor/AP-1 pathway: potential origin for the cell-specific effects of tamoxifen and toremifene on bone biochemistry and bone estrogen-like effects of antiestrogens. Mol. Endocrinol., 9: 443-456, mineral density in postmenopausal breast cancer patients. J. Clin. En- 1995. docrinol. Metab., 83:1158-1162, 1998. 25. Paech, K., Webb, P., Kuiper, G. G., et al. Differential ligand 45. Vogel, C. L., Shemano, I., Schoenfelder, J., et al. R. Multicenter activation of estrogen receptors ER a and ER [3 at AP1 sites. Science Phase II efficacy trial of toremifene in tamoxifen-refractory patients (Wash. DC), 277: 1508-1510, 1997. with advanced breast cancer. J. Clin. Oncol., 11: 345-350, 1993. 26. Schiff, R., Reddy, P., Coronado, E., et al. Development of tamox- 46. Pyrhonen, S., Valavaara, R., Vuorinen, J., et al. High dose ifen-stimulated growth in vivo is associated with changes in AP-1 toremifene in advanced breast cancer resistant to or relapsed during activity. Breast Cancer Res. Treat., 46: A347, 1997. tamoxifen treatment. Breast Cancer Res. Treat., 29: 223-228, 1994. 27. Johnston, S. R. D., Lu, B., Scot, G. K., et al. Increased activator 47. Asaishi, K., Tominaga, T., Abe, O., et aL Efficacy and safety of protein-I DNA binding and c-Jun NH2-terminal kinase activity in hu- high-dose NK 622 (toremifene citrate) in tamoxifen failed patients with man breast tumors with acquired tamoxifen resistance. Clin. Cancer breast cancer. Gan To Kagaku Ryoho, 20: 91-99, 1993. Res., 5: 251-256, 1999. 48. Jonsson, P. E., Malmberg, M., Bergljung, L., et aL Phase II study of 28. Smith, C. L., Nawaz, Z., O'Malley, B. W., et al. Coactivator and high dose toremifene in advanced breast cancer progressing during co-repressor regulation of the agonist/antagonist activity of the mixed tamoxifen treatment. Anticancer Res., 11: 873- 876, 1991.

49. Stenbygaard, L. E., Herrstedt, J., Thomsen, J. F. et al. Toremifene 70. Dhingra, K. Antiestrogens: tamoxifen, SERMs and beyond. Inves- and tamoxifen in advanced breast cancer--a double-blind cross-over fig. New Drugs, 17: 285-311, 1999. trial. Breast Cancer Res. Treat., 25: 57-63, 1993. 71. Lien, E. A., and Lonning, P. E. Selective oestrogen receptor mod- 50. Wiseman, L. R., and Goa, K. L. Toremifene: a review of its ifiers (SERMs) and breast cancer therapy. Cancer Treat. Rev., 26: pharmacological properties and clinical efficacy in the management of 205-227, 2000. advanced breast cancer. Drugs, 54: 141-160. 1997. 72. McCague, R., Leclerq, G., Legros, N., et al. Derivatives of tamox- 51. Valavaara, R., and Pyrhonen, S. Low dose toremifene in the treat- ifen: dependence of antiestrogenicity on the 4-substituent. J. Med. ment of estrogen receptor positive advanced breast cancer in postmeno- Chem., 32: 2527-2533, 1989. pausal women. Curr. Ther. Res., 46: 966-973, 1989. 73. Chander, S. K., McCague, R., Luqmani, Y., et al. Pyrrolidino-4- 52. Nomura, Y., Tominaga, T., Abe, O., et al. Clinical evaluation of NK iodotamoxifen and 4-iodotamoxifen, new analogues of the antiestrogen 622 (toremifene citrate) in advanced or recurrent breast cancer--a tamoxifen for the treatment of breast cancer. Cancer Res., 51: 5851- comparative study by a double-blind method with tamoxifen. Jpn. J. 5858, 1991. Cancer Chemother., 20: 247-258, 1993. 74. Nuttall, M. E., Bradbeer, J. N., Stroup, G. B., et al. A novel 53. Hayes, D. F., Van Zyl, J. A., Hacking, A., et al. Randomised selective estrogen receptor modulator prevents bone loss and lowers comparison of tamoxifen and two separate doses of toremifene in cholesterol in ovariectomised rats and decreases uterine weight in intact postmenopausal patients with metastatic breast cancer. J. Clin. Oncol., rats. Endocrinology, 139: 5224-5234, 1998. 13: 2556-2566, 1995. 75. Johnston, S. R. D., Riddler, S., Haynes, B. P. et al. The novel 54. Pyrhonen, S., Valavaara, R., Modig, H., et aL Comparison of antioestrogen idoxifene inhibits the growth of human MCF-7 breast toremifene and tamoxifen in post-menopausal patients with advanced cancer xenografts and reduces the frequency of acquired antiestrogen breast cancer: a randomised double-blind, the "Nordic" Phase II study. resistance. Br. J. Cancer, 75: 804-809, 1997. Br. J. Cancer, 76: 270-277, 1997. 76. Coombes, R. C., Haynes, B. P., Dowsett, M., et al. Idoxifene: report 55. Gershanovich, M., Garin, A., Baltina, D., et aL A Phase III com- of a Phase I study in patients with metastatic breast cancer. Cancer Res., parison of two toremifene doses to tamoxifen in postmenopausal women 55: 1070-1074, 1995. with advanced breast cancer. Breast Cancer Res. Treat., 45: 251-262, 77. Johnston, S. R. D., Gumbrell, L., Evans, T. R. J., et al. A Phase II 1997. randomized double-blind study of idoxifene (40 mg/d) vs. tamoxifen (40 56. Milla-Santos, A., Milla, L., Rallo, L., et aL Phase III trial of mg/d) in patients with locally advanced/metastatic breast cancer resist- toremifene vs tamoxifen in hormone dependent advanced breast cancer. ant to tamoxifen (20 mg/d). Proc. Am. Soc. Clin. Oncol., 18: 109, 1999. Breast Cancer Res. Treat., 65:119-124, 2001. 78. Johnston, S. R. D., Gorbunova, V., Lichinister, M., et al. A multi- 57. Pyrhonen, S., Ellman, J., Vuorinen, J., et al. Meta-analysis of trials centre double-blind randomised Phase III trial of idoxifene versus ta- comparing toremifene with tamoxifen and factors predicting outcome of moxifen as first-line endocrine therapy for metastatic breast cancer. antiestrogen therapy in postmenopausal women with breast cancer. Proc. Am. Soc. Clin. Oncol., 20: 29, 2001. Breast Cancer Res. Treat., 56: 133-143, 1999. 79. Toko, T., Sugimoto, Y., Matsuo, E., et al. TAT-59, a new triph- 58. Holli, K., Valvaara, R., Blanco, G., et al. Toxicity and early survival enylethylene derivative with antitumour activity against hormone- results of a prospective randomized adjuvant trial comparing toremifene dependent tumors. Eur. J. Cancer, 26: 397-404, 1990. and tamoxifen in node-positive breast cancer. Proc. Am. Soc. Clin. Oncol., 19: 87, 2000. 80. Lino, Y., Takai, Y., Ando, T., et al. A new triphenylethylene derivative TAT-59: hormone receptors, insulin-like growth factor 1 and 59. Roos, W. K., Oeze, L., Loser, R., et al. Antiestrogen action of 3-hydroxy-tamoxifen in the human breast cancer cell line MCF-7. growth suppression of hormone dependent MCF-7 tumours in athymic mice. Cancer Chemother. Pharmacol., 34: 372-376, 1994. J. Natl. Cancer Inst. (Bethesda), 71: 55-59, 1983. 60. Eppenberger, U., Wosikowski, K., and Kung, W. Pharmacologic 81. Wilson, T. M., Norris, J. D., Wagner, B. L., et aL Dissection of the and biologic properties of droloxifene, a new antiestrogen. Am. J. Clin. molecular mechanism of action of GW5638, a novel estrogen receptor Oncol., 141: $5-S14, 1991. ligand, provides insights into the role of estrogen receptor in bone. Endocrinology, 138: 3901-3911, 1997. 61. Loser, R., Seibel, K., Roos, W., et al. In vivo and in vitro antiestrogenic action of 3-hydroxytamoxifen, tamoxifen and 4-hydroxytamox- 82. Ke, H. Z., Paralkar, V. M., Grasser, W. A., et al. Effects of ifen. Eur. J. Cancer Clin. Oncol., 21: 985-990, 1985. CP-336,156, a new non-steroidal estrogen agonist/antagonist, on bone, 62. Hasmann, M., Rattel, B., and Loser, R. Preclinical data for drolox- serum cholesterol, uterus and body composition in rat models. Endocri- nology, 139: 2068-2076, 1998. ifene. Cancer Lett., 84: 89-95, 1994. 63. Ke, H., Simmons, H. A., Pierie, C. M., et aL Droloxifene, a new 83. Howell, A., DeFriend, D., Robertson, J., et al. Response to a estrogen antagonist/agonist, prevents bone loss in ovariectomised rats. specific antiestrogen (ICI182,780) in tamoxifen-resistant breast cancer. Endocrinology, 136: 2435-2441, 1995. Lancet, 345: 29-30, 1995. 64. Stamm, H., Roth, R., Almendral, A., et aL Tolerance and efficacy 84. Robertson, J. F. R., Howell, A., DeFriend, D., et aL Duration of of the antiestrogen droloxifene inpatients with advanced breast cancer. remission to ICI182,780 compared to megestrol acetate in tamoxifen- J. Steroid Biochem., 28 (Suppl.): 1085, 1987. resistant breast cancer. Breast, 6: 86-189, 1997. 65. Brietbach, G. P., Moous, V., Bastert, G., et al. Droloxifene: efficacy 85. Black, L. J., Jones, C. D., and Folcone, J. F. Antagonism of estrogen and endocrine effects in treatment of metastatic breast cancer. J. Steroid action with a new benzothiophene-derived antiestrogen. Life Sci., 32: Biochem., 28 (Suppl.): 1095, 1987. 1031-1036, 1983. 66. Haarstad, H., Gundersen, S., Wist, E., et al. Droloxifene--a new 86. Poulin, R., Merand, Y., Poirier, D. Antiestrogenic properties of antiestrogen. Acta Oncologica, 31: 425-428, 1992. keoxifene, trans-4-hydroxytamoxifen and ICI 164,384, a new steroidal 67. Bruning, P. F. Droloxifene, a new antioestrogen in postmenopausal antiestrogen in ZR-75-1 human breast cancer cells. Breast Cancer Res. advanced breast cancer: preliminary results of a double-blind dose- Treat., 14: 65-76, 1989. finding Phase II trial. Eur. J. Cancer, 28A: 1404-1407, 1992. 87. Gottardis, M. M., and Jordan, V. C. The antitumor action of keox- 68. Rausching, W., and Pritchard, K. I. Droloxifene, a new antiestro- ifene (raloxifene) and tamoxifen in the N-nitromethylurea-induce rat gen: its role in metastatic breast cancer. Breast Cancer Res. Treat., 31: mammary carcinoma model. Cancer Res., 47: 4020-4024, 1987. 83-94, 1994. 88. Balfour, J. A., and Goa, K. L. Raloxifene. Drugs Aging, 12: 69. Haarstad, H., Lonning, P., Gundersen, S., et al. Influence of drolox- 335-341, 1998. ifene on metastatic breast cancer as first-line endocrine treatment. Acta 89. Delmas, P. D., Bjarnason, N. H., Mitlak, B. H., et aL Effects of Oncologica, 37: 365-368, 1998. raloxifene on bone mineral density, serum cholesterol concentrations,

and uterine endometrium in postmenopausal women. N. Engl. J. Meal., 105. Couillard, S., Gutman, M., Labrie, C., et al. Comparison of the 337: 1641-1647, 1997. effects of the antiestrogens EM-800 and tamoxifen on the growth of 90. Buzdar, A., Marcus, C., Holmes, F., et aL Phase II evaluation of human breast ZR-75-1 cancer xenografts in nude mice. Cancer Res., 58: LY156758 in metastatic breast cancer. Oncology, 45: 344-345, 1988. 60-64, 1998. 106. Luo, S., Martel, C., Sourla, A., et al. Comparative effects of 91. Gradishar, W. J., Glusman, J. E., Vogel, C. L., et al. Raloxifene HCL, a new endocrine agent, is active in estrogen receptor positive 28-day treatment with the new anti-estrogen EM-800 and tamoxifen on estrogen-sensitive parameters in intact mice. Int. J. Cancer, 73: 381- (ER +) metastatic breast cancer. Breast Cancer Res. Treat., 46: 53, 1997. 391, 1997. 92. Cummings, S. R., Eckert, S., Krueger, K. A., et al. The effect of 107. Simard, J., Sanchez, R., Poirier, D., et al. Blockade of the stimu- raloxifene on risk of breast cancer in postmenopausal women: results latory effect of estrogens, OH-tamoxifen, OH-toremifene, droloxifene, from the multiple outcomes of raloxifene evaluation (MORE) random- and raloxifene on alkaline phosphatase activity by the antiestrogen ized trial. JAMA, 281: 2189-2197, 1999. EM-800 in human endometrial adenocarcinoma Ishikawa cells. Cancer 93. Cauley, J. A., Norton, L., Lippman, M. E., et al. Continued breast Res., 57: 3494-3497, 1997. cancer risk reduction in postmenopausal women treated with raloxifene: 108. Johnston, S. R. D., Detre, S., Riddler, S., et al. SCH 57068 is a 4-year results from the MORE trial. Breast Cancer Res. Treat., 65: selective estrogen receptor modulator (SERM) without uterotrophic 125-134, 2001. effects compared with either tamoxifen or raloxifene. Breast Cancer 94. Sato, M., Turner, C. H., Wang, T. Y., etal. LY353381.HCI: a novel Res. Treat., A163, 2000. raloxifene analogue with improved SERM potency and efficacy in vivo. 109. Martel, C., Picard, S., Richard, V., et al. Prevention of bone loss J. Pharmacol. Exp. Ther., 287: 1-7, 1998. by EM-800 and raloxifene in the ovariectomised rat. J. Steroid Biochem. 95. Fuchs-Young, R., Iversen, P., Shelton, P., et al. Preclinical demon- Mol. Biol., 74: 45-56, 2000. stration of specific and potent inhibition of mammary tumour growth by 110. Labrie, F., Champagne, P., Labrie, C., et aL Response to the orally new selective estrogen receptor modulators (SERMs). Proc. Am. Assoc. active specific antiestrogen EM-800 (SCH-57070) in tamoxifen-resist- Cancer Res., 38: 573, 1997. ant breast cancer. Breast Cancer Res. Treat., 46: 53, 1997. 96. Johnston, S. R. D., Riddler, S., Detre, S., et al. Dose-dependent 111. Greenberger, L., Komm, B., Miller, C., et al. Pre-clinical pharma- inhibition of MCF-7 breast cancer xenograft growth with LY353381, a cology profile of a new selective estrogen receptor modulator (SERM), novel selective estrogen receptor modulator (SERM). Proc. Am. Assoc. ERA-923, for the treatment of ER-positive breast cancer. Breast Cancer Cancer Res., 41: 25, 2000. Res. Treat., 64: 52, 2000. 97. Sato, M., Zeng, G. Q., Rowley, E., et al. LY353381.HCL: an 112. Nabholtz, J. M., Buzdar, A., Pollack, M., et al. Anastrozole is improved benzothiophene analogue with bone efficacy complimentary superior to tamoxit'en as first-line therapy for advanced breast cancer in to parathyroid hormone-(1-34). Endocrinology, 139:4642-4651, 1998. post-menopausal women: results of a North American multicenter ran- 98. Rowley, E., Adrian, M. D., Bryant, H., et al. The new SERM domized trial. J. Clin. Oncol., 18: 3758-3767, 2000. LY353381.HCL has advantages over estrogen, tamoxifen, and ralox- 113. Bonneterre, J., Thurlimann, B., Robertson, J. F. R., et al. Anas- ifene in reproductive and non-reproductive tissues of aged ovariecto- trozole versus tamoxifen as first-line therapy for advanced breast cancer mised rats. Proc. Am. Soc. Bone Miner. Res., A490, 1997. in 668 postmenopausal women: results of the Tamoxifen or Arimidex 99. Munster, P. N., Buzdar, A., Dhingra, K., et al. Phase I study of a Randomized Group Efficacy and Tolerability Study. J. Clin. Oncol., 18: third generation selective estrogen receptor modulator LY353381.HCL 3748-3757, 2000. in metastatic breast cancer. J. Clin. Oncol., 19: 2002-2009, 2001. 114. Mouridsen, H., Gershanovich, M., Sun, Y., et al. Superior efficacy 100. Baselga, J., Llombart-Cussat, A., Bellet, M., et al. Double-blind of letrozole versus tamoxifen as first-line therapy for postmenopausal randomised Phase II study of a selective estrogen receptor modulator in women with advanced breast cancer: results of a Phase III study of the patients with locally advanced or metastatic breast cancer. Breast Can- International Letrozole Breast Cancer Group. J. Clin. Oncol., 19: 2596- cer Res. Treat., 57: A25, 1999. 2606, 2001. 101. Buzdar, A., O'Shaughnessy, J., Hudis, C., et al. Preliminary re- 115. Masamura, S., Santner, S. J., Heitjan, D. F., et al. Estrogen suits of a randomised double-blind Phase II study of the selective deprivation causes estradiol hypersensitivity in human breast cancer estrogen receptor modulator (SERM) arzoxifene in patients with locally cells. J. Clin. Endocrinol. Metab., 80: 2918-2925, 1995. advanced or metastatic breast cancer. Proc. Am. Soc. Clin. Oncol., 20: 116. Klijn, J. G. N., Blamet, R. W., Boccardo, F., et al. Combined 45, 2001. tamoxifen and luteinizing hormone-releasing hormone (LHRH) agonist 102. Labrie, F., Labrie, C., Belanger, A., et al. EM-652 (SCH 57068), versus LHRH agonist alone in premenopausal advanced breast cancer: a a third-generation SERM acting as a pure antiestrogen in the mammary meta-analysis of four randomized trials. J. Clin. Oncol., 19: 343-353, gland and endometrium. J. Steroid Biochem. Mol. Biol., 69: 51-84, 2001. 1999. 117. Dowsett, M., Dixon, J. M., Horgan, K., et al. Antiproliferative 103. Martel, C., Provencher, L., Li, X., et al. Binding characteristics of effect of idoxifene in a placebo-controlled trial in primary human breast novel antiestrogens to the rat uterine estrogen receptors. J. Steroid cancer. Clin. Cancer Res., 6: 2260-2267, 2000. Biochem. Mol. Biol., 64: 199-205, 1998. 118. Dowsett, M., Lu, Y., Hills, M., et al. Effect of raloxifene on Ki67 104. Simard, J., Labrie, C. L., Belanger, A., et al. Characterisation of and apoptosis. Breast Cancer Res. Treat., 57: 31, 1999. the effects of the novel non-steroidal antiestrogen EM-800 on basal and 119. DeFriend, D., Howell, A., Nicholson, R. I., et al. Investigation of estrogen-induced proliferation of T47-D, ZR-75-1 and MCF-7 human a new pure antiestrogen (ICI 182,780) in women with primary breast breast cancer cells in vitro. Int. J. Cancer, 73:104-112, 1997. cancer. Cancer Res., 54: 408-414, 1995.

Downloaded from clincancerres.aacrjournals.org on September 27, 2021. © 2001 American Association for Cancer Research. Endocrine Manipulation in Advanced Breast Cancer: Recent Advances with SERM Therapies

Stephen R. D. Johnston

Clin Cancer Res 2001;7:4376s-4387s.

Updated version Access the most recent version of this article at: http://clincancerres.aacrjournals.org/content/7/12/4376s

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

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

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/7/12/4376s. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.