A Review of Selective Estrogen Receptor Modulators in theTreatment of Breast and Endometrial Cancer

Stephen Chan

The understanding of how estrogen affects differentbody tissues by selective actions on the two subtypes ofestrogen receptors (alpha and beta) has created thepossibility of targeted therapy by the manufacturing ofa group of compounds known as selective estrogenreceptor modulators. The goal of an ideal selectiveestrogen receptor modulator that has all the beneficialeffects of estrogen receptor modulation without ad-verse side effects seems increasingly achievable withimproving drug design. The clinical findings for the newselective estrogen receptor modulator, arzoxifene,which has been shown to be highly active in the treat-ment of advanced breast cancer as well as advancedendometrial cancer, has confirmed the value of selec-tive targeting of the estrogen receptors, and may her-ald a new era in endocrine therapy in clinical oncology.Semin Oncol 29 (suppl 11):129-133. Copyright 2002,Elsevier Science (USA). All rights reserved.

ESTROGEN has a critical role in the growth,development, and maintenance of a range of

tissues because of its physiologic effects via theestrogen receptor (ER). Both ER isoforms (ERalpha and beta) function as ligand-activated tran-scription factors. The discovery of the ER beta in19961 has had a profound effect on our under-standing of estrogen action: it helps to explainhow estrogen may have a wide range of activity indifferent tissues. Many organs express both ERsubtypes, but different cell types within each organmay express different ERs, suggesting that in someinstances ER alpha and ER beta have distinctfunctions.

The action of estrogen can be modulated by agroup of compounds called selective ER modula-tors (SERMs). Because of the distribution andlocalization of different receptors, the action of agiven SERM depends on its activation of a differ-ent hormone-binding domain, which determinesan agonist, antagonist, or mixed response. The“ideal SERM” is a compound that has an anties-trogenic effect in breast and uterine cancer, buthas a positive estrogenic effect in the skeletal andcardiovascular systems.

This brief review will show how far selectivetargeting of the ER has taken us in an attempt toachieve the ideal SERM for clinical benefit topatients with cancer. These SERMs include (1)faslodex, a pure ER antagonist; (2) tamoxifen andraloxifene (ER agonists in bone), shown to be

effective for the prevention of breast cancer; and(3) arzoxifene, a third-generation SERM that ishighly active for the treatment of advanced breastand endometrial carcinoma.

A PURE ANTIESTROGEN: FASLODEX

The antagonistic effects of SERMs appear toarise from a ligand-induced suboptimal conforma-tion of ER that cannot communicate with thecellular transcriptional apparatus. Faslodex is apure antiestrogen, which leads to downregulationand loss of the ER2,3 and results in the completeabrogation of the ER function. In vitro, faslodexinhibits the growth of tamoxifen-resistant celllines and increases the response duration in ahuman breast tumor model.4 Furthermore, in asmall phase II study of 19 postmenopausal womenwith advanced breast cancer who had receivedprior tamoxifen treatment, faslodex showed signif-icant antitumor efficacy. The partial response ratewas 37% (no complete responses), with 32%achieving stable disease for more than 24 weeks.5The result of a phase III trial comparing faslodexwith standard treatment (tamoxifen) in a first-lineadvanced breast cancer setting may be availablewithin the next 12 months. However, the use offaslodex as adjuvant therapy for early breast cancerand prevention may be limited by its lack of es-trogenic effects in the skeletal and cardiovascularsystems.

SELECTIVE ESTROGEN RECEPTORMODULATORS FOR THE PREVENTION OF

BREAST CANCER: TAMOXIFEN ANDRALOXIFENE

Evidence from breast cancer treatment trialsshowed the original SERM, tamoxifen, to be ef-fective in preventing the development of a second

From the Nottingham City Hospital, Nottingham, United King-dom.

Address reprint requests to Stephen Chan, DM, MRCP, FRCR,Nottingham City Hospital, Hucknall Rd, Nottingham, UK NG51PB.

Copyright 2002, Elsevier Science (USA). All rights reserved.0093-7754/02/2903-1117$35.00/0doi:10.1053/sonc.2002.34065

129Seminars in Oncology, Vol 29, No 3, Suppl 11 (June), 2002: pp 129-133

primary breast cancer in women being treated inthe adjuvant setting for early breast cancer.6 Therole of tamoxifen in the prevention of breast can-cer was confirmed by the National Surgical Adju-vant Breast and Bowel Project Trial (NSABP P-1:BCPT).7 With a minimum of 69 months follow-up, the trial showed that tamoxifen significantlyreduced the risk of breast cancer (primarily inER-positive patients) by 49% with a cumulativeincidence of 172 versus 89 cases of cancer in theplacebo and treatment groups, respectively. Themajor toxicity of tamoxifen in the trial was anincreased risk of endometrial cancer, ie, a relativerisk of 2.53 compared with the control group (95%confidence interval [CI]: 1.35 to 4.97). Two sim-ilar prevention studies in Europe showed less sig-nificant results regarding the efficacy of tamoxifenin chemoprevention.8,9 A fourth trial, the Inter-national Breast Cancer Intervention Trial, whichstarted in 1992, completed recruitment of over5,000 subjects in the year 2000.10

A benzothiophene SERM, raloxifene, with es-trogen agonist effects in bone has been shown asclinically useful for the prevention and treatmentof postmenopausal osteoporosis.11,12 Its estrogenicagonist effects on the metabolism of lipid13 havealso been investigated for the prevention of heartdisease.14 The effect of raloxifene on breast cancerwas a secondary endpoint in the Multiple Out-comes of Raloxifene Evaluation trial. The 4-yearresults from the 7,705 postmenopausal women en-rolled in the study (Fig 1) showed a 72% reductionof breast cancer risk in the raloxifene group com-pared with the placebo group: relative risk 0.28(95% CI, 0.17 to 0.46). The 61 breast cancersreported have been confirmed by an independentpanel.15 As expected, the main effect of raloxifene

is a reduction of risk in patients with ER-positivebreast cancer (84%). Thromboembolic disease oc-curred significantly more frequently with ralox-ifene than placebo (P � .003). Incidences were1.44, 3.32, and 3.63 events per 1,000 woman/yearsfor placebo, raloxifene 60 mg, and raloxifene120 mg/d, respectively. The magnitude of this es-trogenic effect was similar to that seen with ta-moxifen in the NSABP trial. Unlike tamoxifen,there was no increase in endometrial cancer in theraloxifene group; the incidence rates were 0.77,0.77, and 0.60 cases per 1,000 women/years forplacebo, raloxifene 60 mg, and raloxifene120 mg/d, respectively.

A Study of Tamoxifen And Raloxifene, whichcompares these two SERMs in the role of chemo-prevention in a phase III trial setting, started re-cruitment in North America in July 1999. Theplan is to randomize 22,000 postmenopausalwomen to tamoxifen 20 mg or raloxifene 60 mg/dfor 5 years. It will be interesting to see from theresults if the efficacy of SERMs can be improved bybetter targeting of the ER.

A SELECTIVE ESTROGEN RECEPTORMODULATOR FOR THE TREATMENT OFADVANCED BREAST AND ENDOMETRIAL

CANCER: ARZOXIFENE

Arzoxifene is a SERM that acts as a potentestrogen antagonist in mammary and uterine tis-sues while acting as an estrogen agonist to main-tain bone density and lower serum cholesterol.16,17

It is metabolized to a desmethyl metabolite. Boththe parent compound and the metabolite bind tothe ER with high affinity and inhibit estrogen-dependent growth of MCF-7 breast cancercells.18,19 The potency is 3,000- to 25,000-foldhigher than that of tamoxifen against the MCF-7cells. In addition, arzoxifene acts as an estrogenantagonist in uterine tissue. In a rat model that hasbeen used to study the physiology of explantedendometrial tissue, treatment with arzoxifene 10and 30 mg/kg caused a significant reduction in thegrowth of explanted uterine tissue.20 In a phase Imultiple-dose study in patients with metastaticbreast or endometrial cancer, arzoxifene was usedat dose levels of 10, 20, 50, and 100 mg/d.21 Themost frequent toxicity was hot flashes observed atall dose levels in the 32 patients treated. Prelimi-nary pharmacokinetic analyses showed the half-life of unconjugated arzoxifene to be approxi-mately 30 hours. The plasma concentration of the

Fig 1. Effect of raloxifene on breast cancer incidence in the48-month Multiple Outcomes of Raloxifene Evaluation trial.Reprinted with permission.15

130 STEPHEN CHAN

drug was also found to increase linearly and pro-portionately over this dose range.

Advanced Breast Cancer

In a phase II randomized, double-blind study(JWWC), arzoxifene was used as a first-line treat-ment for advanced breast cancer.22 Ninety-foursubjects were randomized to one of two treatmentgroups, 20 or 50 mg/d of arzoxifene, until diseaseprogression. The subject characteristics in botharms were similar and are summarized in Table 1.There were 46 patients treated in each group; twopatients withdrew from the trial before receivingtreatment. All patients had good performance sta-tus. About 90% were postmenopausal, and 75%were ER- or progesterone receptor-positive; 9% ineach group had prior tamoxifen. The efficacy re-sults of this study are summarized in Table 2.Tumor assessment was independently reviewed.Response rates were similar in the two dose levelsat 40.5% and 36.4%, respectively. The clinicalbenefit rate was defined as the sum of objectivetumor response (complete response � partial re-sponse) and stable disease lasting at least 6months. This was similar in the two groups at 64%and 61.4%. In summary, arzoxifene was highlyeffective in the treatment of advanced breast can-cer. The clinical significance of this activity needsto be assessed against the gold standard of tamox-ifen. A phase III study comparing these two drugs

for the treatment of advanced breast cancer is inprogress.

Advanced Endometrial Cancer

In a phase II trial of arzoxifene in advancedendometrial cancer (JWWH), patients with ER/progesterone receptor-positive tumors received20 mg/d of the trial medication until disease pro-gression.23 Patients with unknown receptor statuswere entered provided their tumors were well ormoderately differentiated with a remission dura-tion of over 3 years, or had previously responded toprogesterone. Patients were divided into two co-horts: progesterone sensitive or progesterone fail-ure. Progesterone failure was defined as progressionon progesterone therapy, progression within 8weeks of stopping adjuvant therapy, or discontin-uation of progesterone because of toxicity. Proges-terone sensitive was defined as no prior progester-one or adjuvant therapy with a disease-freeduration of at least 6 months. Table 3 summarizesthe characteristics of patients and Table 4 summa-rizes the tumor characteristics of patients in thestudy. Median age was 69 years, and the majorityof patients had not received prior systemic therapyin the form of progesterone or cytotoxic chemo-therapy. Ten percent of tumor receptors statuswere unknown.

The primary objective of the study was to de-termine antitumor efficacy in terms of tumor re-sponse rate. Table 5 summarizes the overall tumorresponse rate in the 32 evaluable progesterone-

Table 1. Patient Demographics in a RandomizedPhase II Trial of Arzoxifene in Advanced

Breast Cancer22

Arzoxifene

20 mg/d 50 mg/d

Patients randomized 46 46Age, median 70 yrs 69 yrsZubrod performance status

0 26 (57%) 32 (70%)1 20 (43%) 14 (30%)

Menopausal statusPre 4 (9%) 2 (4%)Post 41 (89%) 44 (96%)

ER-progesterone receptor statusPositive (any) 34 (74%) 34 (74%)

ER-progesterone receptorUnknown 10 (22%) 12 (26%)

Prior tamoxifen 4 (9%) 4 (9%)Prior adjuvant chemotherapy 7 (15%) 9 (20%)

Table 2. Efficacy Results of a Randomized Phase IITrial of Arzoxifene in Advanced Breast Cancer22

Investigator

20 mg/d 50 mg/d

Patients randomized 46 46Patients evaluable 44 44Response rate 36.0% 34.0%95% CI (within evaluable

group) 22% to 52% 21% to 50%Clinical benefit rate 64% 64%95% CI (within group) 48% to 78% 48% to 78%Median time to progression

(mos) 10.4 8.995% CI 8% to 13.3% 5.9% to NA

Abbreviations: CI, confidence interval; CR, complete re-sponse; PR, partial response; SD, stable disease; NA, notachieved.

SERMs IN THE TREATMENT OF BREAST CANCER 131

sensitive patients enrolled (28.0%), and the dura-tion of responses ranged from 5.5 months to morethan 12.2 months. With the addition of 11 pa-tients who had stable disease for over 6 months inthe study, the clinical benefit response rate was62.5% (95% CI, not reported). These results aresignificantly higher than those achieved using me-droxyprogesterone acetate in a similar set of pa-tients (who received no prior hormonal therapy)in a recent phase II study conducted by the Gy-necologic Oncology Group,24 in which the re-sponse rate was 25% and median duration of re-sponse was 2 to 3 months. The response rate in amore recent report of a similar phase II study usingtamoxifen was 10% (90% CI, 5.7% to 17.9%).25

There is no doubt that arzoxifene has high anti-tumor activity in advanced endometrial cancer.The clinical role of the drug in this setting canonly be assessed in a phase III study in comparisonagainst the standard therapy of progesterone.

CONCLUSION

Ongoing research into drug design combined witha better understanding of cellular biology has enabledmore specific targeting with SERMs. This in turn has

increased the therapeutic ratio of these drugs, conse-quently yielding additional treatment options forbreast cancer. Examples include faslodex in the treat-ment of tamoxifen-resistant advanced disease5; re-duced risk of uterine endometrial cancer with ralox-ifene (compared with tamoxifen) in the field ofchemoprevention12; and arzoxifene, which may playa major role in the treatment of endometrial cancer.Selective estrogen receptor modulators are also likelyto play an important role in combination with otheragents to regulate the growth of cancer cells becausethere are significant interactions of the ERs andother pathways for therapeutic targeting. The futureof SERMs looks increasingly exciting, with moderncompounds designed to give more specific effects.

REFERENCES

1. Kuiper GG, Enmark E, Pelto-Huikko M, et al: Cloning ofa novel receptor expressed in rat prostate and ovary. Proc NatlAcad Sci U S A 93:5925-5930, 1996

2. Howell A, Downey S, Anderson E: New endocrine ther-apy for breast cancer. Eur J Cancer 32A:576-588, 1996

3. Robertson JF: Faslodex (ICI 182,780), a novel estrogenreceptor downregulator–Future possibilities in breast cancer. JSteroid Biochem Mol Biol 79:209-212, 2001

4. Osborne CK, Coronado-Heinsohn EB, Hilsenbeck SG:Comparison of the effects of a pure steroidal antiestrogen withthose of tamoxifen in a model of human breast cancer. J NatlCancer Inst 87:746-750, 1995

5. Howell A, DeFriend D, Robertson J, et al: Response to aspecific antioestrogen (ICI 182780) in tamoxifen-resistantbreast cancer. Lancet 345:29-30, 1995

6. Nayfield SG, Karp JE, Ford LG, et al: Potential role oftamoxifen in prevention of breast cancer. J Natl Cancer Inst83:1450-1459, 1991

Table 3. Patient Demographics of a Phase II Trial ofArzoxifene in Advanced Endometrial Cancer23

Median age 69 yearsPrior radiation therapy 53.0%Prior chemotherapy 8.0%

Table 4. Tumor Characteristics of Patients in a PhaseII Study of Arzoxifene in Advanced

Endometrial Cancer23

Receptor statusER�/PgR� 81.8%ER�/PgR unknown 2.7%ER�/PgR� 5.4%ER�/PgR� 5.4%ER/PgR unknown 5.4%

Grade of endometrial cancerGrade 1 32.4%Grade 2 51.4%Grade 3 16.2%

Abbreviations: ER, estrogen receptor; PgR, progesteronereceptor.

Table 5. Investigator-Assessed Tumor Response in aPhase II Trial of Arzoxifene in Advanced Endometrial

Cancer: Progesterone-Sensitive Patients23

Investigator-Assessed(n � 32)

Objective tumor response (CR �

PR) 1 � 8Response rate 28.0%95% CI (within group) 14.0% to 47.0%Clinical benefit response

(CR � PR � SD � 6 mos) 1 � 8 � 11Clinical benefit rate* 62.5%Median time to progression (mos) 5.595% CI 3.6% to 9.3%

Abbreviations: CR, complete response; PR, partial response;SD, stable disease; CI, confidence interval.

*CI not reported.

132 STEPHEN CHAN

7. Fisher B, Costantino JP, Wickerham DL, et al: Tamoxifenfor prevention of breast cancer: Report of the National SurgicalAdjuvant Breast and Bowel Project P-1 study. J Natl CancerInst 90:1371-1388, 1998

8. Veronesi U, Maisonneuve P, Costa A, et al: Prevention ofbreast cancer with tamoxifen: Preliminary findings from theItalian randomised trial among hysterectomised women. ItalianTamoxifen Prevention Study. Lancet 352:93-97, 1998

9. Powles TJ, Eeles R, Ashley S, et al: Interim analysis of theincidence of breast cancer in the Royal Marsden Hospitaltamoxifen randomised chemoprevention trial. Lancet 352:98-101, 1998

10. Cuzick J, Howell A: A brief review of the breast cancerprevention trials. Eur J Cancer 36:S51-S53, 2000 (suppl 4)

11. Delmas PD, Bjarnason NH, Mitlak BH, et al: Effects ofraloxifene on bone mineral density, serum cholesterol concen-trations, and uterine endometrium in postmenopausal women.N Engl J Med 337:1641-1647, 1997

12. Ettinger B, Black DM, Mitak BH, et al: Reduction ofvertebral fracture risk in postmenopausal women with osteopo-rosis treated with raloxifene: Results from a 3-year randomizedclinical trial. Multiple Outcomes of Raloxifene Evaluation(MORE) Investigators. JAMA 282:637-645, 1999

13. Walsh BW, Kuller LH, Wild RA, et al: Effects of ralox-ifen on serum lipids and coagulation factors in healthy post-menopausal women. JAMA 279:1445-1451, 1998

14. Barrett-Connor E, Wenger NK, Grady D, et al: Hor-mone and nonhormone therapy for the maintenance of post-menopausal health: The need of randomized controlled trials ofestrogen and raloxifene. J Womens Health 7:839-847, 1998

15. Cauley JA, Norton L, Lippman ME, et al: Continuedbreast cancer risk reduction in postmenopausal women treatedwith raloxifene: 4-year results from the MORE trial. MultipleOutcomes of Raloxifene Evaluation. Breast Cancer Res Treat65:125-134, 2001

16. Cole HW, Adrian MD, Shelter PK, et al: Comparativepharmacology of high potency selective estrogen modulators(SERMS): LY353381. HCI and CP336,156. Proc Am SocBone Mineral Res 12:A491, 1997 (suppl 1) (abstr)

17. Rowley E, Adrian MD, Bryant H, et al: The new SERMLY353381oHCI has advantages over estrogen, tamoxifen andraloxifene in reproductive and nonreproductive tissues of agedovariectomized rats. Proc Am Soc Bone Mineral Res 12:A490,1997 (suppl 1) (abstr)

18. Rash T, Knadler MP: The disposition and biotransfor-mation of the selective estrogen receptor for modulator,LY353381, in female Fischer 344 rats following single oral dose.Proc Am Assoc Pharmaceut Scientists. Pharm Res 14:A1223,1997 (abstr)

19. Lantz R, Knadler MP: The disposition of the selectiveestrogen receptor modulator, LY353381, and its metabolites infemale Cynomolgus monkeys following single oral and intra-venous doses of 4C-LY353381. Proc Am Assoc PharmaceutScientists. Pharm Res 14:A1222, 1997 (abstr)

20. Vernon MW: Experimental endometriosis in laboratoryanimals as a research model. Prog Clin Biol Res 323:49-60, 1990

21. Munster PN, Buzdar A, Dhingra K, et al: Phase I studyof a third-generation selective estrogen receptor modulator,LY353381.HCI, in metastatic breast cancer. J Clin Oncol 19:2002-2009, 2001

22. Llombart-Cussac A, Bellet M, Guillem-Porta V, et al:Efficacy and safety of two doses of the selective estrogen receptormodulator (Serm) Ly353381 in locally advanced or metastaticbreast cancer (Lambc)—A randomized double blind phase 2study. Proc Am Soc Clin Oncol 19:157a, 2000 (abstr 609O)

23. Klijn J, den Hoed Kliniek D: Multicentre phase II studyof selective estrogen receptor modulator (SERM) LY353381 inadvanced or recurrent endometrial cancer–Objective responsesin progestagen sensitive patients. Proc Am Soc Clin Oncol19:386a, 2000 (abstr 1527)

24. Thigpen JT, Brady MF, Alvarez RD, et al: Oral medroxy-progesterone acetate in the treatment of advanced or recurrentendometrial carcinoma: A dose-response study by the Gyneco-logic Oncology Group. J Clin Oncol 17:1736-1744, 1999

25. Thigpen T, Brady MF, Homesley HD, et al: Tamoxifenin the treatment of advanced or recurrent endometrial carci-noma: A Gynecologic Oncology Group study. J Clin Oncol19:364-367, 2001

SERMs IN THE TREATMENT OF BREAST CANCER 133