Aromatase Inhibition and Antiestrogen Therapy in Early Breast Cancer Treatment and Chemoprevention

ABSTRACT: The aromatase inhibitors represent an important class ofhormonal agents for the management of breast cancer. The third-generationaromatase inhibitors have replaced megestrol acetate as second-line hormonaltherapy in advanced breast cancer, and large clinical trials are maturing toestablish their efficacy relative to tamoxifen (Nolvadex) in the first-linemetastatic setting. The increased potency, increased specificity, andestablished efficacy of aromatase inhibitors in advanced breast cancer haveprovided the rationale for a large number of randomized trials in the adjuvantsetting evaluating anastrozole (Arimidex), exemestane (Aromasin), and letrozole(Femara). These trials are addressing the value of these agents in sequencewith, instead of, and in combination with tamoxifen. The relationship betweenestrogen exposure and breast cancer risk has long been accepted andtraditionally related to estrogen-receptor-mediated events. The emergence ofthe estrogen genotoxicity hypothesis as a mechanism for breast cancercarcinogenesis provides additional rationale for considering aromataseinhibitors in the chemoprevention setting. [ONCOLOGY 15(Suppl 7):28-34, 2001]

Introduction

Hormonal therapy remains an integral part of the managementof breast cancer for most women. Tamoxifen (Nolvadex) has been the hormonalagent of choice for women with estrogen-receptor- and/or progesterone-receptor-positive tumors in both the adjuvant and metastatic disease settings.It has also demonstrated efficacy in the chemoprevention setting. Althoughtamoxifen has had a profound and positive impact on the management of breastcancer, the availability of the newer anti-aromatase agents offers the potentialfor improving the therapeutic index in women with breast cancer. In addition,these agents offer a new approach for consideration in chemoprevention of thedisease.

Estrogen and Breast Cancer

Estrogen exposure has long been known to be a risk factor forbreast cancer development. Early menarche, late menopause, and obesity in thepostmenopausal setting (which is associated with high estrogen levels) are allassociated with greater risk of breast cancer. Particularly striking in thisregard are the data on surgically induced menopause from Trichopoulos et al,[1]who found a 60% reduction in breast cancer risk in women so treated comparedwith those experiencing a natural menopause. Although they found the reductiongreatest for women whose menopause was induced before age 35, reductions werealso seen in women up to the age of 50 years.

Interference with estrogen action and reduction in estrogenproduction are two major approaches taken in the management of hormone-dependentbreast cancer. The interference with estrogen action is best exemplified by theuse of tamoxifen, where the mechanism of action is thought to be primarilyrelated to competition with estrogen for the estrogen receptor, which results indecreased expression of estrogen-regulated genes.[2] Reduction in estrogenproduction is exemplified by the use of ovarian ablation in premenopausal womenand use of agents that inhibit the aromatase system in postmenopausal women.

Converting to Estrogen

In postmenopausal or castrated women, the major source ofestrogen is the peripheral conversion of androstenedione, which is secreted bythe adrenal gland, via aromatase, which is a complex of a cytochrome P450hemoprotein and a flavoprotein, NADD (nicotinamide adenine dinucleotidediphosphate) cytochrome P450 reductase. Aromatization is the rate-limiting stepin the conversion of androstenedione to estrone, which can subsequently bereduced to estradiol.

Adipose tissue contains substantial aromatase activity and isresponsible for much of the peripheral conversion of androstenedione to estrone.The breast, however, is also a site for this peripheral conversion, with obviousimplications for breast cancer development. Miller and O’Neill[3] examinedaromatase activity in different quadrants of mastectomy specimens and found theactivity was always higher in the quadrant associated with the tumor.

High Estradiol Levels

The concentration of estradiol in plasma is substantially lowerin post-menopausal than in premenopausal women. Despite this differential,Pasqualini et al[4] found that the levels of estradiol in breast cancer tissuewere similar in premenopausal and postmenopausal women, and that the levels ofestrone sulfate were significantly higher in the latter patients. Esteban etal[5], using immunohistochemical methodology, found that 15 (40%) of 38 breastcancers demonstrated significant immunoreactivity for aromatase. Utilizing anassay for aromatase activity that quantifies production of tritiated waterrelease from 1-tritiated- androstenedione, Lipton et al[6] found measurablearomatase activity in 69% of 113 breast cancers.

The finding of relatively high estradiol levels in breastcancers in postmenopausal women and the identification of aromatase activityraises the question of the relative importance of in situ aromatization vsenhanced uptake from plasma. An intriguing study addressing this issue wasperformed by Yue et al,[7] who developed a model in which aromatase- andsham-transfected MCF-7 cells were inoculated into ovariectomized nude mice andthen androstenedione was administered. Heterotransplants in whicharomatase-transfected MCF-7 cells were implanted in one flank andsham-transfected cells in the other flank demonstrated that the tumor weight was7.6-fold larger and the estradiol concentration 3- to 4-fold higher in thearomatase-transfected tumors.

Yue also examined the impact of estradiol administration bysilastic implants and androstenedione by injection in ovariectomized mice withimplantation of aromatase-transfected MCF-7 cells. The estradiol administrationalways produced lower tumor growth rates and lower estradiol levels thanandrostenedione, supporting the position that in situ aromatization plays amajor role in determining the tissue estradiol levels.[7] An additionalconsideration is that, whereas circulating estrogens may be bound to sex hormonebinding globulins, this would not be the case in estrogens produced in situ andacting in a paracrine or intracrine fashion.

Aromatase Inhibitors

The three generations of aromatase inhibitors are listed in Table 1. Steroidal aromatase inhibitors, also known as type I, interact with thesubstrate binding site on the peptidic moiety of the enzyme complex in anirreversible manner, which has led to their designation as aromataseinactivators. Nonsteroidal aromatase inhibitors, also known as type II,coordinate with the iron atom of the cytochrome P450 moiety of the enzyme in areversible manner. The striking change that has occurred with the advent of thethird-generation aromatase inhibitors is the increased potency and increasedspecificity, with activity confined to aromatase inhibition, over earliergenerations.[8] Additionally, these agents have markedly improved tolerability.Of particular importance is that the clinical antitumor efficacy has been shownto be improved with third-generation agents.

Gershanovich et al[9] performed a prospective randomized trialin postmenopausal women with advanced breast cancer comparing letrozole (Femara)with aminoglutethimide (Cytadren), and found that letrozole resulted in superiortime to disease progression, time to treatment failure, and overall survival,and was associated with fewer side effects.

Large randomized phase III trials comparing megestrol acetatewith each of the third-generation aromatase inhibitors resulted in replacingmegestrol acetate (Megace) as the treatment of choice in women with advancedbreast cancer who have failed tamoxifen.[10-13]

Aromatase Inhibitors vs Tamoxifen

First- and Second-Generation Inhibitors

Several, albeit relatively small, randomized clinical trialshave been performed comparing first- and second-generation aromatase inhibitorswith tamoxifen as first-line hormonal therapy in the advanced breast cancersetting. In a study by Smith et al,[14] 117 patients were randomly assigned toreceive either tamoxifen or aminoglutethimide, and the two treatments wereassociated with equal objective response rates (30%) and median durations ofresponse (15 months). Although toxicity was greater with aminoglutethimide, thelevel of efficacy appeared equal between the two agents.

Formestane (4-OHA, 4-hydroxy-androstenedione) was compared withtamoxifen in 348 evaluable patients with objective response rates of 33% and37%, median durations of response of 15 and 20 months, and overall survivals of35 and 38 months, respectively.[15] However, time to disease progression andtime to treatment failure significantly favored tamoxifen.

Fadrozole (CGS 16949A) was compared with tamoxifen in 212evaluable patients. Response rates were 20% and 27%, respectively, and meantimes to treatment failure were 6.1 and 8.5 months, respectively (P = .09).[16]Thus, although aromatase inhibitors were shown to have activity, tamoxifenremained the hormonal agent of choice in the first-line setting.

Third-Generation Inhibitors

Information relating to the value of third-generation aromataseinhibitors relative to tamoxifen is beginning to appear; the most mature datarelate to anastrozole (Arimidex). Buzdar et al[17] recently presented theresults of two randomized double-blind studies (T27, T30) that comparedanastrozole and tamoxifen as first-line therapy for advanced breast cancer. Thislarge study involved 1,021 patients who were estrogen-receptor positive and/orprogesterone-receptor positive, or who had an unknown receptor status. Thecomplete response and partial response rates were similar (anastrozole 29%,tamoxifen 27%), with median times to treatment failure of 8.5 and 7.0 months,respectively.

Of particular interest is that the progression-hazard ratio hada lower bound of 1.0, indicating that anastrozole was at least equivalent totamoxifen in this regard. In subset analysis, in the 57.7% of patients known tobe hormone-receptor positive, a significantly longer time to progression wasidentified for the anastrozole group. From a toxicity standpoint, fewerthromboembolic events and vaginal bleeding were seen with anastrozole.

Other Aromatase Inhibitors

Less information is available with the other aromataseinhibitors. A large randomized study involving more than 900 patients comparingletrozole with tamoxifen has completed accrual. Results should be forthcoming. Aphase III study of exemestane (Aromasin) vs tamoxifen is being conducted as anextension of a randomized phase II trial[18] that had demonstrated encouragingresults regarding exemestane efficacy.

Aromatase Inhibitors in the Adjuvant Setting

Tamoxifen is the gold standard against which all new adjuvanthormonal therapies in postmenopausal women must be compared. Based on theNational Surgical Adjuvant Breast and Bowel Project’s trial B14, the currentconsensus regarding duration of tamoxifen treatment is 5 years.[19] At thisduration of therapy, the worldwide overview found that tamoxifen producedsignificant and substantial proportional risk reductions in disease recurrenceand death—50% and 28%, respectively—in women who were estrogen-receptorpositive.[20]

Very few data are available evaluating aromatase inhibitors inthe adjuvant setting in early breast cancer. Jones et al[21] reported a study of336 assessable postmenopausal patients with resected node-positive breastcancer. This randomized double blind, placebo-controlled trial evaluatedaminoglutethimide (250 mg qid) plus hydrocortisone (20 mg daily) for 2 years.There was no benefit with the administration of aminoglutethimide in terms ofevent-free survival (P = .41) or overall survival (P = .098). In a subsetanalysis of patients who were estrogen-receptor positive, the event-freesurvival for aminoglutethimide was superior and approached statisticalsignificance (P = .054), but overall survival was not significantly different.This trial suffered from an inadequate sample size and a high proportion ofpatients (43%) without estrogen-receptor data.

Recently, Boccardo et al[22] presented early results of a trialin which patients who had received approximately 3 years of tamoxifen were thenrandomly assigned to receive an additional 2 years of either tamoxifen oraminoglutethimide. In 380 patients randomized from September 1992 throughJanuary 1998, there was no difference (P = .8) in event-free survival, butoverall survival was significantly better (P = .005) for the sequential tamoxifen followed by the aminoglutethimide arm.This latter arm had more patients relapsing with local disease and fewerrelapsing with visceral disease than the tamoxifen-alone arm. Many of the issuesraised by this small trial should be adequately addressed by the ongoing trialsnoted below.

The Value of Aromatase Inhibitors

Although there is a paucity of data with the earlier aromataseinhibitors, a wealth of data will be forthcoming relating to thethird-generation aromatase inhibitors in the adjuvant setting, as can be seen bythe listing of trials in Table 2. The value of aromatase inhibitors can beconsidered in three categories: (1) their use in sequence with tamoxifen, (2)their use instead of tamoxifen, and (3) their use in combination with tamoxifen.

In Sequence With Tamoxifen

The use of aromatase inhibitors in sequence with tamoxifen isattractive, as tamoxifen is clearly shown to be of value in the worldwideoverview.[20] In considering the duration of tamoxifen treatment, recent datafrom a large Swedish trial are of note. This trial randomized 4,183 patientstreated with tamoxifen for either 2 or 5 years. A significant improvement inboth event-free survival and overall survival was demonstrated for the longertreatment duration.[23]

Despite the superiority of 5 years of tamoxifen, several trialsare addressing shorter duration of this agent followed by aromatase inhibitors.The use of aromatase inhibitors after 5 years of tamoxifen is particularlyattractive because of the lack of potential for stimulating cancer cell growthand the potential for further benefit against any remaining malignant cells.Studies from several of these clinical trials groups, which will provideinformation regarding sequential use of aromatase inhibitors with tamoxifen inthe adjuvant setting, are listed in Table 2.

Instead of Tamoxifen

The use of aromatase inhibitors instead of tamoxifen iscurrently under investigation in two large trials by the Cancer ResearchCampaign’s Breast Cancer Trial Group (the Arimidex, Tamoxifen, Alone or inCombination [ATAC] trial) and the International Breast Cancer Study Group (Table2). Of particular note is the ATAC trial, which is a double-blind trial that hascompleted accrual of more than 9,000 patients. These trials should provide ananswer regarding the value of this approach.

In Combination With Tamoxifen

The use of aromatase inhibitors in combination with tamoxifen isattractive because their mechanisms of action would appear to be complementary.That is, tamoxifen acts as a competitive inhibitor with estradiol for theestrogen receptor, and the reduction of estradiol levels by aromatase inhibitorswould be expected to allow tamoxifen to compete more effectively.

To date, phase III trials in metastatic disease have beenconducted comparing tamoxifen plus aminoglutethimide with tamoxifen alone, andsuperiority for the combination has not been identified.[24] Lien et al[25]subsequently demonstrated a pharmacokinetic interaction between the two agents,which results in a substantial reduction in tamoxifen and its major metabolites.This apparent induction of tamoxifen metabolism provides a potential explanationfor failure of the combination to show superiority.

Preclinical studies have addressed the question of combiningthird-generation aromatase inhibitors with tamoxifen. Lu et al[26] injectedMCF-7 cells transfected with the aromatase gene into ovariectomized nude micethat were administered androstenedione. Neither combination of anastrozole plustamoxifen or letrozole plus tamoxifen was superior to the respective aromataseinhibitor alone. The combination of exemestane and tamoxifen was examined byZaccheo et al[27] in rats with DMBA-induced mammary tumors. In these studies,the use of tamoxifen and exemestane alone appeared equally effective, but thecombination was superior. The ATAC trial (see Table2) is testing the use ofanastrozole plus tamoxifen in the adjuvant setting and will address theimportant issue of combination hormonal therapy utilizing a third-generationnonsteroidal aromatase inhibitor.

Chemoprevention ofBreast Cancer

Tamoxifen

Reduction in breast cancer risk became a clinical reality withthe results of the National Surgical Adjuvant Breast Project (NSABP) P-1trial.[28] This randomized, placebo-controlled trial evaluated the use oftamoxifen for 5 years in more than 13,000 women at high risk (defined as a5-year predicted risk for developing breast cancer of at least 1.66%). Tamoxifenreduced the risk of invasive breast cancer by 49%. This reduction was seen inall age groups, ie, 44% in those ≤ 49 years old, 51% in those aged 50 to 59,and 55% in those 60 years or older. Also, on the positive side, tamoxifenreduced the rate of hip, Colles’, and spine fractures.

On the negative side, tamoxifen was associated with increasedrisk of endometrial cancers and thromboembolic phenomena of stroke, pulmonaryemboli, and deep venous thrombosis. Although two smaller trials[29,30] did notcorroborate the value of tamoxifen in reducing risk of breast cancerdevelopment, the NSABP P-1 results were sufficient to persuade the US Food andDrug Administration to approve tamoxifen for this purpose. It is against thisbackground that discussions of chemoprevention in the clinical setting must beheld.

Aromatase Inhibitors

Similar data in the clinical setting are not available foraromatase inhibitors. Data are available, however, in the preclinical setting.Gunson et al[31] evaluated the second-generation aromatase inhibitor fadrozolein female Sprague-Dawley rats, which have ahigh tendency for spontaneous development of breast cancer. A dose-dependentprevention of breast cancer development was identified. Lubet et al[32] studiedthe third-generation aromatase inhibitor vorozole (Rivizor) in femaleSprague-Dawley rats given methylnitrosourea and, likewise,found a dose-dependent inhibition of mammary cancers. There is burgeoninginterest in the breast cancer chemoprevention potential for aromataseinhibitors.[33-35]

Estrogen-Related Carcinogenesis

A commonly held view on the role of estrogen in breast cancerdevelopment is that the effects are mediated through the estrogen receptor inthe breast cancer cell. Thus, the more exposure of breast cells to estrogen, themore proliferation and the greater potential for mutations that could lead tomalignancy. Recently, increased attention has been paid to the potential forestrogen to function as a cancer-inducing agent through metabolism to genotoxicmetabolites. The two different pathways are presented in Figure1. It isentirely possible that both of these pathways are operational.

Genotoxicity

In the course of oxidative metabolism of estrogen, semiquinonesand quinones are formed, which can result in the formation of either stable ordepurinating DNA adducts.[36] Additional potential for DNA damage can come fromactive oxygen species that are generated in this process. It is clear from thepreceding discussion that the potential for genotoxic estrogen metabolitesexists, which represents a potential mechanism for breast cancer carcinogenesis.

Numerous enzymes are involved in the metabolic activation andinactivation of estrogen, and genetic polymorphisms have been described for manyof these enzymes. One example is that of catechol O-methyl transferase, forwhich the polymorphism was described in 1977[37] and characterized at themolecular level in 1996.[38] These investigators identified that thispolymorphism has almost equal frequencies for an allele that encodes amethionine (which is associated with low activity) or valine (which isassociated with high activity) at a specific condon.

Of particular interest are three reports that have indicatedthat individuals who are homozygous for the low activity allele are at increasedrisk for breast cancer development.[39-41] This single example raises thepossibility that estrogen genotoxicity may play a role in breast cancerdevelopment.

The aromatase enzyme is obviously central to the issue ofestrogen genotoxicity in postmenopausal women, as it is responsible for theconversionof androstenedione to estrogen; pertinent in this regard are recent reports ofpolymorphisms in the human aromatase gene. Siegelmann-Danieli and Buetow[42]identified an association between a specific polymorphism and the high-riskgenotype. Kristensen et al[43] found a specific polymorphic allele of CYP19 tobe significantly more frequent in patients with breast cancer than in controls.These two examples raise the possibility that in the future, high-risk genotypescould be identified in which aromatase inhibitors could be examined aschemopreventive agents.

Prevention Possibilities

Inhibition of aromatase to prevent breast cancer would appear tobe a hypothesis worthy of testing. Several issues need to be considered whichare, in reality, analogous to those operative in the adjuvant setting.

Similar to the adjuvant setting, tamoxifen is the gold standardin the chemoprevention setting. Tamoxifen, and possibly other selectiveestrogen-receptor modulators, will need to be considered in the testingstrategy. However, the negative aspects of tamoxifen therapy, such asendometrial cancers and thromboembolic phenomena, represent opportunities forthe aromatase inhibitors, as these toxicities would not be expected with theseagents.

A concern with aromatase inhibitors is the potential for adverseeffects on bone density and on lipid profiles; these concerns are beingaddressed in clinical research at present. Fortunately, nonhormonal measures areavailable that could be combined with aromatase inhibitors to counteract theiradverse effects on bone and lipids.

Conclusion

The aromatase inhibitors/inactivators are an important class ofagents that are assuming a greater role in the clinical management of breastcancer. There is a strong scientific rationale for their use, not only in thesetting of adjuvant therapy but also in the chemoprevention setting. Numerouslarge randomized clinical trials are being conducted to address the value ofthese agents in the adjuvant setting. Extremely valuable information related toreduction of contralateral breast cancers will be available from these studies.Additional information regarding the value of aromatase inhibitors in theadjuvant setting will be necessary prior to development of large-scale clinicaltrials for breast cancer chemoprevention.

Note added in proof: Data have recently been presented from arandomized double-blind trial comparing letrozole with tamoxifen in 907postmenopausal women with locally advanced or metastatic breast cancer.[44]Letrozole was superior to tamoxifen in terms of time to progression (medians: 41vs 26 weeks, P = .0001), objective response rate (complete plus partialresponse: 30% vs 20%, P = .001), and clinical benefit rate (complete pluspartial response plus stable disease for at least 24 weeks: 49% vs 38%, P =.001). The results from this trial provided the basis for FDA approval ofletrozole in the first-line treatment of advanced breast cancer.

References:

1. Trichopoulos D, MacMahon B, Cole P: Menopause and breastcancer risk. J Natl Cancer Inst 48:605-613, 1972.

2. Osborne CK: Tamoxifen in the treatment of breast cancer. NEngl J Med 339:1609-1618, 1998.

3. Miller WR, O’Neill J: The importance of local synthesis ofestrogen within the breast. Steroids 50:537-548, 1987.

4. Pasqualini JR, Chetrite G, Blacker C, et al: Concentrationsof estrone, estradiol, and estrone sulfate and evaluation of sulfatase andaromatase activities in pre- and postmenopausal breast cancer patients. J ClinEndocrinol Metab 81:1460-1464, 1996.

5. Esteban JM, Warsi Z, Haniu M, et al: Detection ofintratumoral aromatase in breast carcinomas. Am J Pathol 140:337-343, 1992.

6. Lipton A, Santen RJ, Santner SJ, et al: Prognostic value ofbreast cancer aromatase. Cancer 70:1951-1955, 1992.

7. Yue W, Wang J-P, Hamilton CJ, et al: In situ aromatizationenhances breast tumor estradiol levels and cellular proliferation. Cancer Res58:927-932, 1998.

8. Santen RJ, Harvey HA: Use of aromatase inhibitors in breastcarcinoma. Endocr Relat Cancer 6:75-92, 1999.

9. Gershanovich M, Chaudri HA, Campos D, et al: Letrozole, a neworal aromatase inhibitor: Randomised trial comparing 2.5 mg daily, 0.5 mg dailyand aminoglutethimide in postmenopausal women with advanced breast cancer. AnnOncol 9:639-645, 1998.

10. Buzdar AU, Jonat W, Howell A, et al: Anastrozole versusmegestrol acetate in the treatment of postmenopausal women with advanced breastcarcinoma. Results of a survival update based on a combined analysis of datafrom two mature phase III trials. Cancer 83:1142-1152, 1998.

11. Dombernowsky P, Smith I, Falkson G, et al: Letrozole, a neworal aromatase inhibitor for advanced breast cancer: Double-blind randomizedtrial showing a dose effect and improved efficacy and tolerability compared withmegestrol acetate. J Clin Oncol 16:453-461, 1998.

12. Goss PE, Winer EP, Tannock IF, et al: Randomized phase IIItrial comparing the new potent and selective third-generation aromataseinhibitor vorozole with megestrol acetate in postmenopausal advanced breastcancer patients. J Clin Oncol 17:52-63, 1999.

13. Kaufmann M, Bajetta E, Dirix LY, et al: Exemestane issuperior to megestrol acetate after tamoxifen failure in postmenopausal womenwith advanced breast cancer: Results of a phase III randomized double-blindtrial. J Clin Oncol 18:1399-1411, 2000.

14. Smith IE, Harris AL, Morgan M, et al: Tamoxifen versusaminoglutethimide in advanced breast carcinoma: A randomised cross-over trial.Br Med J 283:1432-1434, 1981.

15. Perez Carrion R, Candel VA, Calabresi F, et al: Comparisonof the selective aromatase inhibitor formestane with tamoxifen as first-linehormonal therapy in postmenopausal women with advanced breast cancer. Ann Oncol5(suppl 7):S19-S24, 1994.

16. Thürlimann B, Beretta K, Bacchi M, et al: First-linefadrozole HCl (CGS 16949A) versus tamoxifen in postmenopausal women withadvanced breast cancer. Ann Oncol 7:471-479, 1996.

17. Buzdar A, Nabholtz JM, Robertson JF, et al: Anastrozole(Arimidex) versus tamoxifen as first-line therapy for advanced breast cancer inpostmenopausal women—Combined analysis from two identically designedmulticenter trials. Proc Am Soc Clin Oncol 19:154a, 2000.

18. Paridaens R, Dirix LY, Beex L, et al: Exemestane (Aromasin)is active and well tolerated as first-line hormonal therapy of metastatic breastcancer patients: Results of a randomized phase II trial. Proc Am Soc Clin Oncol19:83a, 2000.

19. Fisher B, Dignam J, Bryant J, et al: Five versus more thanfive years of tamoxifen therapy for breast cancer patients with negative lymphnodes and estrogen receptor-positive tumors. J Natl Cancer Inst 88:1529-1542,1996.

20. Early Breast Cancer Trialists’ Collaborative Group:Tamoxifen for early breast cancer: An overview of the randomised trials. Lancet351:1451-1467, 1998.

21. Jones AL, Powles TJ, Law M, et al: Adjuvantaminoglutethimide for postmenopausal patients with primary breast cancer:Analysis at 8 years. J Clin Oncol 10:1547-1552, 1992.

22. Boccardo F, Rubagotti A, Amoroso D, et al: Tamoxifen vsaminoglutethimide in breast cancer patients previously treated with adjuvantTAM. Preliminary results of a multicentric comparative study. Proc Am Soc ClinOncol 19:71a, 2000.

23. Nordenskjold B, Carstensen J, Rutqvist L, et al: Prolongedfollow-up of the Swedish randomized trial of two versus five years of adjuvanttamoxifen for postmenopausal early stage breast cancer and relationships tohormone receptor and ERBB2 levels. Proc Am Soc Clin Oncol 19:72a, 2000.

24. Ingle JN, Suman VJ, Jordan VC, et al: Combination hormonaltherapy involving aromatase inhibitors in the management of women with breastcancer. Endocr Relat Cancer 6:265-269, 1999.

25. Lien EA, Anker G, Lønning PE, et al: Decreased serumconcentrations of tamoxifen and its metabolites induced by aminoglutethimide.Cancer Res 50:5851-5857, 1990.

26. Lu Q, Liu Y, Long BJ, et al: The effect of combiningaromatase inhibitors with antiestrogens on tumor growth in a nude mouse modelfor breast cancer. Breast Cancer Res Treat 57:183-192, 1999.

27. Zaccheo T, Giudici D, DiSalle E: Inhibitory effect ofcombined treatment with the aromatase inhibitor exemestane and tamoxifen onDMBA-induced mammary tumors in rats. J Steroid Biochem Mol Biol 44:677-680,1993.

28. Fisher B, Costantino JP, Wickerham DL, et al: Tamoxifen forprevention of breast cancer: Report of the National Surgical Adjuvant Breast andBowel Project P-1 study. J Natl Cancer Inst 90:1371-1388, 1998.

29. Veronesi U, Maisonneuve P, Costa A, et al: Prevention ofbreast cancer with tamoxifen: Preliminary findings from the Italian randomisedtrial among hysterectomised women. Lancet 352:93-97, 1998.

30. Powles T, Eeles R, Ashley S, et al: Interim analysis of theincidence of breast cancer in the Royal Marsden Hospital tamoxifen randomisedchemoprevention trial. Lancet 352:98-101, 1998.

31. Gunson DE, Steele RE, Chau RY: Prevention of spontaneoustumours in female rats by fadrozole hydrochloride, an aromatase inhibitor. Br JCancer 72:72-75, 1995.

32. Lubet RA, Steele VE, DeCoster R,et al: Chemopreventive effects of the aromatase inhibitor vorozole (R 83842) inthe methylnitrosourea-induced mammary cancer model. Carcinogenesis 19:1345-1351,1998.

33. Kelloff GJ, Lubet RA, Lieberman R, et al: Aromataseinhibitors as potential cancer chemopreventives. Cancer Epidemiol BiomarkersPrev 7:67-78, 1998.

34. Santen RJ, Yue W, Naftolin F, et al: The potential ofaromatase inhibitors in breast cancer prevention. Endocr Relat Cancer 6:235-243,1999.

35. Dowsett M: Theoretical considerations for the idealaromatase inhibitor. Breast Cancer Res Treat 49:S39-S44, 1998.

36. Cavalieri EL, Stack DE, Devanesan PD, et al: Molecularorigin of cancer: Catechol estrogen-3,4-quinones as endogenous tumor initiators.Proc Natl Acad Sci USA 94:10937-10942, 1997.

37. Weinshilboum RM, Raymond FA: Inheritance of low erythrocytecatechol-o-methyltransferase activity in man. Am J Hum Genet 29:125-135, 1977.

38. Lachman HM, Papolos DF, Saito T, et al: Humancatechol-O-methyltransferase pharmacogenetics: Description of a functionalpolymorphism and its potential application to neuropsychiatric disorders.Pharmacogenetics 6:243-250, 1996.

39. Lavigne JA, Helzlsouer KJ, Huang H-Y, et al: An associationbetween the allele coding for a low activity variant ofcatechol-O-methyltransferase and the risk for breast cancer. Cancer Res57:5493-5497, 1997.

40. Thompson PA, Shields PG, Freudenheim JL, et al: Geneticpolymorphisms in catechol-O-methyltransferase, menopausal status, and breastcancer risk. Cancer Res 58:2107-2110, 1998.

41. Huang CS, Chern HD, Chang KJ, et al: Breast cancer riskassociated with genotype polymorphism of the estrogen-metabolizing genes CYP17,CYP1A1, and COMT: A multigenic study on cancer susceptibility. Cancer Res59:4870-4875, 1999.

42. Siegelmann-Danieli N, Buetow KH: Constitutional geneticvariation at the human aromatase gene (Cyp19) and breast cancer risk. Br JCancer 79:456-463, 1999.

43. Kristensen VN, Harada N, Yoshimura N, et al: Geneticvariants of CYP19 (aromatase) and breast cancer risk. Oncogene 1919:1329-1333,2000.

44. Smith R, Sun Y, Garin A, et al: Femara (letrozole) showedsignificant improvement in efficacy over tamoxifen as first-line treatment inpostmenopausal women with advanced breast cancer (abstract 7). Breast Cancer ResTreat 64:27, 2000.