Aromatase (estrogen synthetase) inhibitors are superior to tamoxifen in terms of both efficacy and toxicity in the treatment of advanced breast cancer and also in the neoadjuvant setting. Recent results from the Arimidex, Tamoxifen, Alone or in Combination adjuvant trial showed a marked reduction in contralateral primary breast cancer with anastrozole, an apparent prevention effect. A similar effect was seen in the MA.17 adjuvant trial comparing letrozole with placebo after 5 years of adjuvant tamoxifen. This has accelerated interest in aromatase inhibitors as primary preventive therapy. Two studies being conducted by the National Cancer Institute of Canada’s Clinical Trials Group select women by virtue of mammographic breast density. The International Breast Cancer Intervention Study 2 trial randomizes women at elevated risk to anastrozole or placebo. Because of its steroidal structure, exemestane may be more effective than the nonsteroidal aromatase inhibitors and may protect bone and lipid metabolism from the effects of estrogen ablation. Elevated prostaglandin E2 levels from cyclooxygenase-2 induction by preinvasive and invasive breast lesions increase a number of tumor-promoting pathways, including aromatase, as well as angiogenetic, antiapoptotic, and others. Additive or synergistic effects between celecoxib, a cyclooxygenase-2 inhibitor, and exemestane have been demonstrated and have led to the National Cancer Institute of Canada’s Clinical Trials Group MAP.3 trial, which will randomize women at elevated risk to placebo or to exemestane with or without celecoxib. The efficacy and long-term toxicity data from the aromatase inhibitor prevention trials, and the identification of risk profiles from trial results, are awaited with interest.

Epidemiological data have for some time indicated a hormonally mediated basis for breast cancer. Women of advancing age most commonly get breast cancer. Early menarche, late menopause, and nulliparity, all markers of prolonged ovulatory function, are associated with an increase in risk. Use of prolonged hormone replacement therapy increases risk and premature oophorectomy reduces it. Obesity, a marker of increased total body estrogen production, is associated with an increased risk (1).

In vitro and in vivo evidence indicate that estrogen plays a central role in the promotion of breast cancer and possibly in its initiation as well. In vitro, estrogen exerts direct and indirect proliferative effects on breast cancer cells, produces alkylation of cellular molecules, and generates active potentially carcinogenic metabolites, e.g., catechol estrogens and 16 α-hydroxyestrone (2, 3, 4, 5, 6, 7). Estrogens are capable of initiating cancer independent of the estrogen receptor (ER) in the ER knockout mouse (8). Recently, breast cancer patients have been shown to have higher levels of the catechol estrogens 4-OHE2 and 4-OHE1 in their breasts (9).

Clinical markers of cumulative lifelong estrogen exposure have been identified. For example, serum-free estradiol and estrone levels in postmenopausal women have been tightly correlated with subsequent breast cancer risk (10, 11, 12). Similarly, menopausal women in the upper quartile of bone density have increased breast cancer risk, and conversely, women with osteoporosis have a lower than age-matched incidence of breast cancer (13). In the Multiple Outcomes of Raloxifene Evaluation trial in menopausal women with osteoporosis, breast cancer appeared to be tightly linked to estradiol levels, and raloxifene reduced the incidence only in those with elevated levels (14). Breast density on screening mammography has been associated with increased risk, and although one study has failed to demonstrate increased plasma estrogen levels in women with increased density (15), it is possible that this is an estrogen-mediated phenomenon related to elevated intrabreast estrogen levels or increased sensitivity to estrogen.

Taken together, antagonizing estrogen must be viewed as a prime target for breast cancer prevention. The pathogenic pathway to breast cancer is ill-understood but widely believed to be a slow process with a preinvasive phase proceeding from normal epithelium through ductal hyperplasia and then atypical ductal hyperplasia (ADH) and finally ductal carcinoma in situ (DCIS). The exact molecular mechanisms have not as yet been elucidated, but by the time DCIS is present, aneuploidy exists and overexpression of multiple growth-promoting pathways is detectable, including the ER (16, 17), cyclooxygenase-2 (cox-2; Ref. 18), p53 (17), c-ErbB2 (17), and cyclins (17). The ER in particular becomes increasingly expressed, with ∼100% of ADH lesions being positive for ER expression (16). Antagonizing estrogen may therefore reduce the occurrence of the precursor lesions of breast cancer and, in doing so, the occurrence of invasive cancer as well. As breast cancer is one of the leading causes of cancer-related death in women, the long-term goal of any prevention strategy is to reduce mortality from this disease.

As proof of principle, four Phase III trials have been conducted randomizing women at risk of breast cancer to tamoxifen or placebo. Although tamoxifen is known to cause a temporary clinical response of disease in women with advanced, receptor-positive disease, the results are more impressive in early-stage breast cancer, yielding a one-third reduction in mortality 15 years from diagnosis in women receiving 5 years of postoperative therapy (19). Women were selected by different criteria in each of the four tamoxifen trials, resulting in slightly differing risk profiles and therefore potentially different therapeutic indices when comparing efficacy to toxicities. In the National Surgical Adjuvant Breast and Bowel Project P1 trial, women were selected based on a Gail score of >1.66 (20). In the Royal Marsden trial, women with a family history were selected and were allowed administration of hormone replacement therapy for alleviation of symptoms while on study (21). In the International Breast Cancer Intervention Study (IBIS) 1, trial preinvasive histology or family history was used as selection criteria (22), and in the Italian study (23), trial entry was confined to women having had a hysterectomy and concomitant estrogen replacement therapy was allowed.

A recent meta-analysis of the updated results from these trials indicated that all four trials were independently positive and that overall tamoxifen reduced the incidence of breast cancer by 38% (24). Importantly, tamoxifen was shown to reduce the incidence of invasive and preinvasive breast lesions in all age groups; however, the effect appeared to be in reducing receptor-positive tumors only, with receptor-negative tumors being similar in incidence to the control group who received placebo. Toxicities of tamoxifen have been well described from trials in early breast cancer and were confirmed by the meta-analysis of their data (24), as well as from the National Surgical Adjuvant Breast and Bowel Project P1 prevention trial (25). Endometrial cancer and thromboembolic episodes, although rare, were the most serious toxicities, their occurrence virtually undescribed in women < 50 years and most common with increasing age after menopause. Vasomotor and genitourinary adverse effects were the most common and troublesome toxicities, seen at all ages but most frequently in perimenopausal women. Interestingly, all estrogen-dependent benign breast lesions were also markedly reduced, as was the number of breast biopsies among the tamoxifen-treated women (26). The ongoing Study of Tamoxifen and Raloxifene (National Surgical Adjuvant Breast and Bowel Project P2) trial is now comparing tamoxifen to raloxifene in postmenopausal women at risk of breast cancer defined by the same criteria as for the P1 study but excluding premenopausal women. The rationale for this trial is that raloxifene, a Selective Estrogen Receptor Modulator distinct from tamoxifen, produced a substantial reduction in invasive ER-positive breast cancers in the Multiple Outcomes of Raloxifene Evaluation trial (14). The apparent absence of uterine toxicities associated with raloxifene is important, but venous thromboembolism and vasomotor symptoms are likely equally or more troublesome with raloxifene than with tamoxifen.

The third generation aromatase inhibitors, anastrozole, letrozole and exemestane, have been used as monotherapy in postmenopausal women with advanced disease and in ongoing trials of early-stage breast cancer (27, 28).

Anastrozole and letrozole have been approved widely as first-line therapy and exemestane in tamoxifen-refractory disease in postmenopausal women with receptor-positive disease. Randomized trials of anastrozole and letrozole indicated superior efficacy and equivalent short-term toxicities compared with tamoxifen (29, 30, 31).

In the adjuvant setting in early-stage breast cancer, tamoxifen was compared with anastrozole and to the combination of tamoxifen and anastrozole in the Arimidex, Tamoxifen, Alone or in Combination (ATAC) trial (32). Early results from this trial indicated a superior reduction in disease-free survival for anastrozole compared with tamoxifen and in particular a marked reduction in contralateral breast cancer limited apparently to receptor-positive lesions (Fig. 1). Importantly, in the ATAC trial a greater number of long bone fractures were reported for the anastrozole group. In a separate patient-completed quality-of-life assessment, Fallowfield et al.(33) reported loss of interest in sex, vaginal dryness and pain with intercourse more prominently with anastrozole than tamoxifen. In addition, nausea and vomiting were more frequent with the aromatase inhibitor.

Recently, the results of the MA.17 adjuvant study, comparing 5 years of letrozole versus placebo after 5 years of adjuvant tamoxifen were published. This study was discontinued at the first interim analysis because of a highly significant reduction in breast cancer recurrences in the letrozole arm. After a median follow-up of 2.4 years, the hazard ratio for local and distant recurrence or new contralateral breast cancer in the letrozole group as compared with the placebo group was 0.57 (95% confidence interval, 0.43–0.75, P = 0.00008). Low-grade hot flashes, arthritis, arthralgia, and myalgia were more frequent in the letrozole group, but vaginal bleeding was less frequent. Similar to the ATAC trial, there were more new diagnoses of osteoporosis in women in the placebo group, but this was not statistically significant (34).

On the basis of the close association of estrogen with breast cancer as indicated above and the potential to remove not only estrogen-ER interaction but also nonreceptor-based actions of estrogen and its metabolites, aromatase inhibitors provide a logical alternative to tamoxifen as a means of antagonizing the effects of estrogen. In follow-up to the IBIS 1 trial showing tamoxifen to be superior to placebo but showing no net health benefits for tamoxifen, the IBIS group has begun a trial randomizing women to anastrozole or placebo (IBIS 2). Substudies are planned to assess the safety of the inhibitor on bone metabolism and other potential endpoints of estrogen deprivation, including cognitive function. The National Cancer Institute of Canada’s Clinical Trials Group is conducting a 1-year pilot study of letrozole (2.5 mg/day) against placebo in 120 postmenopausal women with a prior history of breast cancer and increased breast density on mammogram. The primary end point will be breast density; careful bone and lipid assessments are included in the study protocol.

The third generation aromatase inhibitors can be classified in several ways (27), one of which is to subdivide them into those with a steroidal chemical structure such as exemestane and those that are nonsteroidal, i.e., letrozole and anastrozole. Exemestane is structurally related to the natural estrogen precursor androstenedione, and this similarity may have implications for both its therapeutic efficacy as well as its toxicities and end-organ effects. Firstly, exemestane binds distinctly and irreversibly to the aromatase enzyme complex (35). In vitro, this binding can be shown to reduce intracellular aromatase content in human cultured breast fibroblasts (35). In contrast, levels tend to rise when the culture system is exposed to the nonsteroidal competitive inhibitors (35). Aromatase inhibition is thought to exert its therapeutic effect by reducing circulating plasma levels of estrogen, but although less well established, additional targets of estrogen synthesis may include premalignant and primary tumor cells, metastatic sites of tumor cells, and probably peritumoral fibroblasts as well (36). It is at these latter targets that differences in inhibition may occur in vivo, potentially resulting in increases in local estrogen production with the nonsteroidal inhibitors, which in turn may circumvent the primary action of the drugs and create disease resistance. In vitro, hormone-dependent breast cancer cells grown in the presence of stable estrogen concentrations should not be growth inhibited. However, with the addition of exemestane to this culture system, proliferation is inhibited, suggesting an additional second mechanism of action independent of estrogen suppression. Moreover, this inhibition can be reversed by the addition of the androgen receptor blocker flutamide, suggesting that this additional antitumor action of exemestane is mediated through the androgen receptor.1

A putative androgenic effect by exemestane given at therapeutic doses to postmenopausal women is also evidenced by a stepwise reduction in sex hormone-binding globulin levels in Phase I studies of the drug. The 10-month-old castrated female Sprague Dawley rat is a typical model for assessing bone metabolism. Castration produces prompt osteoporosis in the femur and the lumbar vertebrae. In two independent experiments conducted in our laboratory, the osteoporosis engendered by castration could be completely alleviated by simultaneous administration of exemestane or its principal rat and human metabolite, 17-hydroexemestane (37, 38). This protective effect was not shown with concurrent administration of letrozole, the nonsteroidal aromatase inhibitor, suggesting that bone metabolism is protected by the androgenicity of exemestane and its metabolite. In both cases the drugs were given at antitumor concentrations in tumor-bearing animals. Interestingly, in our rat experiment, a similar protective effect was seen on the adverse changes in lipid metabolism induced by castration. In ovariectomized animals, exemestane resulted in a 28% inhibition of serum cholesterol levels (P < 0.0001 versus ovariectomized controls) and reduced low-density lipoprotein by 64% compared with ovariectomized controls (P < 0.002; Ref. 37).

In a short-term 3-month postmenopausal human volunteer experiment, exemestane or letrozole was administered daily at the usual anticancer doses of 25 and 2.5 mg, respectively, to each of 20 women, and a placebo was given to 20 women representing a control group (39). Over this short time period, bone biomarkers of formation and resorption, hence net bone turnover, were similar in the women given exemestane or placebo. In contrast, there was a sharp increase in bone turnover in those given letrozole over the same time period (Fig. 2). These findings are compatible with previous data showing increased bone turnover and concomitant reduction in bone mineral density in women receiving letrozole, vorozole and anastrozole (32, 40, 41, 42, 43).

When these data are taken together, it appears that the steroidal inhibitor exemestane may have a superior therapeutic index over the nonsteroidal inhibitors, potentially affording superior efficacy and a better end-organ profile. Androgenic effects of the compound may also produce differences in vasomotor, urogenital, and psychosexual function. If data are confirmed, the profile of exemestane may make it a superior choice to the nonsteroidal agents when contemplated as long-term therapy for prevention in otherwise healthy postmenopausal women. Several large adjuvant trials and smaller studies in healthy high-risk women are currently ongoing and will provide important data on the long term safety of the steroidal inhibitor. The National Cancer Institute of Canada’s Clinical Trials Group is currently conducting a randomized study of exemestane versus placebo in 120 healthy postmenopausal women with increased breast density on screening mammography. Detailed assessments of bone and lipid metabolism are included with the evaluation of changes in mammographic density in this study. The NASBP B-33 trial is evaluating exemestane for 2 years versus placebo after 5 years of adjuvant tamoxifen. The Tamoxifen, Exemestane, Adjuvant Multicenter Trial study compares tamoxifen to exemestane, both for 5 years, in patients with newly diagnosed breast cancer. Within B-33, several substudies will evaluate the effects of the treatment on bone and lipid metabolism, breast density, endometrium, and quality of life. The International Cancer Collaborative Group is conducting a trial in which patients with early-stage breast cancer are randomized to exemestane or tamoxifen for 2–3 years after having completed 2–3 years of adjuvant tamoxifen. The long-term safety data on exemestane reported from these studies will clarify whether the preclinical superiority of exemestane over the other inhibitors can be translated into clinical benefit.

There are epidemiological data suggesting that aspirin and nonsteroidal anti-inflammatory drugs reduce the risk of breast cancer. Recently, corroborative evidence for this was provided by the 43% reduction in breast cancer incidence in women in the Women’s Health Initiative study taking over-the-counter nonsteroidal anti-inflammatory drugs (44).

Cox-2 is a specific inducible pathway shown to be up-regulated in DCIS and in invasive cancers. Induction of cox-2 leads to pleiotropic up-regulation of growth-promoting pathways (45, 46). These include induction of vascular endothelial growth factor, stimulation of ligand binding through the epidermal growth factor receptor, stimulation of an antiapoptotic pathway, and induction of aromatase. In turn, cox-2 may be additionally up-regulated as a positive feedback mechanism by both the epidermal growth factor receptor pathway and by estrogen via ER-β (Fig. 3). Inducible cox-2 is also closely associated with aromatase expression in preinvasive and invasive lesions (47). In animals bearing hormone-dependent breast tumors, the cox-2 inhibitor celecoxib causes a dose-dependent reduction in the incidence of tumors (prevention) and shrinkage of existing tumors (therapy). In the same model, additive or synergistic effects are seen with exemestane and celecoxib on the prevention of tumors in the animals (Fig. 4).

Finally, in vitro celecoxib exerts a dose-dependent antiproliferative effect on the receptor-negative human breast cancer cell line, MDA-MB-231 (48). Celecoxib is an approved anti-inflammatory agent in widespread use clinically and in addition has recently been approved as an anticancer agent for familial polyposis coli. Consequently, celecoxib is of potential interest both alone and in conjunction with an aromatase inhibitor in breast cancer prevention. Pilot trials in pre- and postmenopausal women at risk for breast cancer are under way using celecoxib as a single agent.

The National Cancer Institute of Canada’s Clinical Trials Group is undertaking an international Phase III trial of exemestane with or without celecoxib against a placebo in women at increased risk of breast cancer (Fig. 5). This trial will be the first definitive Phase III trial of breast cancer prevention designed to potentially reduce not only receptor-positive disease but also receptor negative, which contributes disproportionately to mortality from breast cancer. Detailed assessments of bone health, lipid profile, quality of life, cognitive function, and menopausal symptoms will be made within the context of this trial. In addition, plasma and peripheral blood DNA will be collected on all study subjects. The results of this study will be awaited with interest.

It is of interest that the postmenopausal breast contains estrogen concentrations comparable with those found in premenopausal breasts despite plasma levels being ∼10-fold lower (35, 49). Miller et al.(35) have shown that this is in part because of locally synthesized estrogen within the breast itself. Conceivably, therefore, it is possible to reduce intrabreast estrogen levels without necessarily ablating circulating levels. If possible this would have the obvious advantage of antagonizing the effects of estrogen within the breast without the end-organ and symptomatic consequences of complete estrogen blockade. One way of achieving this could be to use a very mild type or dose of aromatase inhibitor. Examples of this strategy could include a small dose of a third generation inhibitor or use of a second generation inhibitor such as 4-hydroxyandrostenedione. Celecoxib itself may be a mild inhibitor of induced aromatase in preinvasive and invasive disease and investigating its use as monotherapy in this regard is merited. Complete ablation of both peripheral and intrabreast aromatase with a third generation inhibitor and adding back a physiological dose of estrogen is another potential way of reducing intrabreast estrogen and retaining circulating estrogen.

In the ATAC study in early-stage breast cancer, a third combination arm of tamoxifen plus anastrozole was studied. Early results from the trial suggested that women receiving the combination had a comparable outcome in terms of disease recurrence compared with those taking tamoxifen alone. The concept behind this trial was total estrogen blockade in the hope of improving the therapeutic efficacy while exploiting the positive effects of tamoxifen on bone and lipid metabolism to counter the putative negative effects of the aromatase inhibitor. Before the ATAC trial, Brodie et al.(50) had used a nude mouse model bearing aromatase up-regulated MCF-7 cells to predict that the combination arm would be less effective than the aromatase inhibitor alone. This was true for both anastrozole and letrozole in this preclinical model (50). The most popular hypothesis for failure of tamoxifen to add to the aromatase inhibitor’s effects is the suggestion that tamoxifen’s agonist effects are provoked or exaggerated in the presence of low estrogen levels. However, most recently Brodie has shown in the same model that exemestane plus tamoxifen is superior to either drug alone (51). This has renewed interest in this concept, and a trial of exemestane and tamoxifen in metastatic disease has been proposed. The original intent of such a combination remains of potential interest in breast cancer prevention. These hypothetical strategies are summarized in Table 1.

Aromatase inhibitors are replacing tamoxifen as first-line treatment of advanced breast cancer and are poised to do so in early-stage disease. The inhibitors appear more effective and appear to lack the serious toxicities of tamoxifen. However, the prolonged therapeutic action of tamoxifen remains to be demonstrated for the inhibitors and careful evaluation of their long-term toxicities is necessary. In particular, the consequences for bone and lipid metabolism and cardiovascular risk need to be determined. In addition, effects on menopausal quality of life and the urogenital tract need to be clarified. Tamoxifen has established the proof of principle for breast cancer prevention, but its therapeutic benefits in healthy women seem superior in younger women < 50 years of age or in those who have had a hysterectomy. In postmenopausal women in whom endometrial cancer risk, thromboembolism and urogenital dysfunction are most pronounced, the aromatase inhibitors may represent a logical alternative to tamoxifen. The American Society of Clinical Oncology technology assessment undertaken in 2003 regarding tamoxifen’s use in prevention concluded that no net health benefit for tamoxifen has been demonstrated to date in healthy women and advocated placebo-controlled trials for future prevention studies (52). In this regard, the IBIS 2 trial of anastrozole against placebo seems appropriate, as does the National Cancer Institute of Canada’s Clinical Trials Group MAP.3 trial of exemestane with or without celecoxib against placebo. The latter study is innovative in as far as it introduces the potentially optimal aromatase inhibitor with androgenic protective effects against end-organs and the potential of preventing receptor-negative disease as well as receptor positive. Strategies of combining aromatase inhibitors with other cell signaling pathway inhibitors and attempting to reduce intrabreast estrogen levels are of additional interest for the future.

Dr. Myles Brown: How do you justify choosing exemestane? I understand you have preclinical data suggesting that antiandrogenic effects may be therapeutic, but are there adjuvant data equivalent to the ATAC trial to suggest a significant prevention effect with exemestane?

Dr. Goss: This goes to the whole discussion on whether the aromatase inhibitors are individual agents or have a class effect. All the metastatic data coincide. Exemestane has been used adjuvantly in about 8,000 women. Our prevention trial is built upon a vanguard of data that is emerging in a timely manner. Coombe’s adjuvant exemestane trial with over 4,000 patients is going to present data in the third quarter of this year, including contralateral breast cancer incidence.

Dr. Brown: You are hopeful that is going to support your study. The other question I have is about your model for prevention of ER-negative disease. It seems to me that the majority of ER-negative tumors cannot have an ER-positive precursor, because otherwise we would have seen at least some inkling in the National Surgical Adjuvant Breast and Bowel Project-P1 trial of preventing them.

Dr. Goss: Perhaps not in the timeframe that we have seen with the trials. If, for instance, ADH converts itself to ER-negative/ER-positive DCIS, which then goes on eventually to form various cancers, it would take a long time to see a reduction in ER negative, if you hypothesize that is the route to ER-negative tumors.

Dr. Brown: A more likely hypothesis is that ADH is the precursor for ER-positive breast cancer that goes to well-differentiated, mostly ER-positive DCIS and that goes to ER-positive cancer. The ER-negative tumors are arising from some other process. There are plenty of ER-negative cells in the breast, and there is no reason why they couldn’t be the precursors to ER-negative tumors.

Dr. Richard Santen: There is a third hypothesis: that estrogen is causing breast cancer independently of the ER. We have put forward a working hypothesis that estradiol is converted to 3,4-quinone, which can be directly genotoxic. If that is the case, one would expect that an aromatase inhibitor would block the development of both ER-negative and ER-positive tumors. We are all anxiously awaiting the findings of the ATAC trial, where they are going to determine the receptor status of the contralateral breast cancers, to see if in fact that is the case.

Dr. Goss: One of the things that intrigues me about breast cancer is the fact that in the people with the highest estrogen levels, the premenopausal women, you get the most ER-negative beast cancer. In postmenopausal women you get few, in men you get almost none. The high estrogen levels in premenopausal women may somehow be responsible for premenopausal women getting ER-negative breast cancer.

Dr. Brown: If Dr. Santen is right, that may be the mutagenic effects of those high estrogen levels. In the postmenopausal women who have up-regulated the receptors, there is a promotional effect in supersensitivity to the low postmenopausal levels of hormones.

Dr. James Ingle: It is important to remember that postmenopausal women have the same amount of estrogen in their breast as premenopausal patients; this concept that you should get a decreased risk of breast cancer as you get older is predicated on the fact that the plasma estrogen levels are different. The breast levels are the same.

Dr. Santen: We don’t know how estrogen levels are maintained in the breast. One possibility is that this represents an endocrine phenomenon where estradiol circulates in the blood and is transported to the breast. Another possibility is local production in the breast through the conversion of androgens to estrogens via local aromatase activity. Finally, both mechanisms may be taking place. We believe that the current data suggest a combination of the two.

Dr. Mina Bissell: Why can’t this be studied in humans, this question of whether or not you have local production or whether this is systemic?

Dr. Ingle: This has been done by Miller et al. at Edinburgh in a very nice study of 31 patients, which they have presented at international meetings (W. R. Miller et al., Vopr Onkol., 47: 182–186, 2001).

Dr. Bissell: Those women already had breast cancer. People often draw conclusions from studies that are done in patients as to how the things are regulated. We know that can be unreliable with cases like TGF-β, which is an inhibitor of breast growth, but in precancerous tissue, it is an inducer of growth; it changes just like that. So the regulation in a noncancerous woman may be very different.

Dr. Santen: Yes, those studies are in patients with breast cancer, and that’s the difficulty. There have been a lot of studies looking at the amount of enzyme and estrogen present in breast cancer tissue and in the distal tissue in a woman who has breast cancer. Now, in women without breast cancer, one can look at the concentrations in the blood and in the breast. These are not all direct studies, but they do suggest that quite a substantial fraction of estrogen is made in the breast itself.

Dr. Ingle: But it really varies substantially from 0 to over 90%. There is a spectrum.

Dr. Bissell: How do you then do all those correlations between plasma level of estrogen and the rates of breast cancer in premenopausal and postmenopausal women, if nobody is measuring breast estrogen, they are all measuring plasma levels?

Dr. Santen: If plasma estrogen is a marker for total body aromatase activity, it might be a marker for intra-breast aromatase also, so it could be a surrogate marker for the amount of estrogen made in the breast. Testosterone, which does correlate quite well with risk of breast cancer, is the substrate for aromatase, and it is possible, then, that its production is local as well. We have to begin to measure estrogen in the breast, and we have to be able to make the appropriate correlations.

Dr. Brown: We need a clinical test that is practical, that measures serum estrogen or some other serum hormone that is correlated with risk, comparable to measuring serum cholesterol. It doesn’t really matter how proximate it is to the cause of the disease; it just has to be able to identify the people at risk.

Dr. Goss: That is why we are prospectively tying together bone density, breast density, and plasma hormone levels, because those are three commonly employed measures that we could subsequently use in trials. I still think that an outcomes trial is needed to validate these measures, and it is going to be hard to do that.

Presented at the Third International Conference on Recent Advances and Future Directions in Endocrine Manipulation of Breast Cancer, July 21–22, 2003, Cambridge, MA.

Grant support: Novartis, Pfizer, Schering.

Requests for reprints: Dr. Paul E. Goss, Professor of Medicine, Director, Breast Cancer Prevention Program, Princess Margaret Hospital, Medical Oncology, Room 5-303, 610 University Avenue, Toronto, Ontario M5G 2M9, Canada. Phone: (416) 946-4501, ext. 5103; Fax: (416) 946-2983; E-mail: pegoss@interlog.com

1

F. Labrie, personal communication.

Fig. 1.

Reduction in new breast primaries with anastrozole compared with tamoxifen and combination.

Fig. 1.

Reduction in new breast primaries with anastrozole compared with tamoxifen and combination.

Close modal
Fig. 2.

Bone biomarker changes at 3 months in healthy postmenopausal controls and in women treated with exemestane 25 or letrozole 2.5 mg daily.

Fig. 2.

Bone biomarker changes at 3 months in healthy postmenopausal controls and in women treated with exemestane 25 or letrozole 2.5 mg daily.

Close modal
Fig. 3.

Biological pathways up-regulated by cycloogenase-2 (cox-2) and relationship to estrogen synthesis (PI3K, phosphoinositide 3′-kinase; MAPK, mitogen-activated protein kinase; P450arom, P450 aromatase).

Fig. 3.

Biological pathways up-regulated by cycloogenase-2 (cox-2) and relationship to estrogen synthesis (PI3K, phosphoinositide 3′-kinase; MAPK, mitogen-activated protein kinase; P450arom, P450 aromatase).

Close modal
Fig. 4.

Celecoxib enhances exemestane tumor growth inhibition in 7,12-dimethylbenz(a)anthracene-induced breast cancer estrogen receptor-positive model. Courtesy of Enrico Pesenti, Discovery Research Oncology, Pharmacia Corporation, Nerviano, Italy.

Fig. 4.

Celecoxib enhances exemestane tumor growth inhibition in 7,12-dimethylbenz(a)anthracene-induced breast cancer estrogen receptor-positive model. Courtesy of Enrico Pesenti, Discovery Research Oncology, Pharmacia Corporation, Nerviano, Italy.

Close modal
Fig. 5.

National Cancer Institute of Canada’s Clinical Trials Group (NCIC CTG) MAP.3 breast cancer prevention trial: exemestane 25 mg daily for 5 years, celecoxib 400 mg daily for 3 years.

Fig. 5.

National Cancer Institute of Canada’s Clinical Trials Group (NCIC CTG) MAP.3 breast cancer prevention trial: exemestane 25 mg daily for 5 years, celecoxib 400 mg daily for 3 years.

Close modal
Table 1

Hypothetical breast cancer prevention strategies: aromatase inhibition (AI)

RegimenPotential effects on peripheral estrogen levelsPotential effects on intrabreast estrogen levels
Full-dose AI ↓↓ ↓↓↓ 
Cycloogenase-2 inhibitor+ AI ↓↓↓ ↓↓↓↓ 
Full-dose AI+ add-back E2 Normal ↓↓ 
Low-dose AI Normal ↓↓ 
Cycloogenase-2 inhibitor Normal ↓ 
RegimenPotential effects on peripheral estrogen levelsPotential effects on intrabreast estrogen levels
Full-dose AI ↓↓ ↓↓↓ 
Cycloogenase-2 inhibitor+ AI ↓↓↓ ↓↓↓↓ 
Full-dose AI+ add-back E2 Normal ↓↓ 
Low-dose AI Normal ↓↓ 
Cycloogenase-2 inhibitor Normal ↓ 
1
Clemons M., Goss P. E. Mechanisms of disease: estrogen and the risk of breast cancer.
N. Engl. J. Med.
,
344
:
276
-285,  
2001
.
2
Harvell D. M., Strecker T. E., Tochacek M., Xie B., Pennington K. L., McComb R. D., Roy S. K., Shull J. D. Rat strain-specific actions of 17β-estradiol in the mammary gland: correlation between estrogen-induced lobuloalveolar hyperplasia and susceptibility to estrogen-induced mammary cancers.
Proc. Natl. Acad. Sci. USA
,
97
:
2779
-2784,  
2000
.
3
Liehr J. G., Fang W. F., Sirbasku D. A., Ari-Ulubelen A. Carcinogenicity of catechol estrogens in Syrian hamsters.
J. Steroid Biochem.
,
24
:
353
-356,  
1986
.
4
Kirkman H. Hormone-related tumors in Syrian hamsters.
Prog. Exp. Tumor Res.
,
16
:
201
-240,  
1972
.
5
Li J. J., Li S. A. Estrogen carcinogenesis in hamster tissues: role of metabolism.
Fed. Proc.
,
46
:
1858
-1863,  
1987
.
6
Kirkman H. Estrogen-induced tumors of the kidney. III. Growth characteristics in the Syrian hamster.
J. Natl. Cancer Inst. Monogr.
,
1
:
1
-57,  
1959
.
7
Newbold R. R., Liehr J. G. Induction of uterine adenocarcinoma in CD-1 mice by catechol estrogens.
Cancer Res.
,
60
:
235
-237,  
2000
.
8
Devanesan P., Santen R. J., Bocchinfuso W. P., Korach K. S., Rogan E. G., Cavalieri E. Catechol estrogen metabolites and conjugates in mammary tumors and hyperplastic tissue from estrogen receptor-α knock-out (ERKO)/Wnt-1 mice: implications for initiation of mammary tumors.
Carcinogenesis (Lond.)
,
22
:
1573
-1576,  
2001
.
9
Kohli E., Zahid M., Rogan E., Cavalieri E. L. Greater reactivity of estradiol-3,4-quinone versus estradiol-2,3-quinone with DNA in the formation of depurinating adducts.
Proc. Am. Assoc. Cancer Res.
,
44
:
180
2003
.
10
Toniolo P. G., Levitz M., Zeleniuch-Jacquotte A., Banerjee S., Koenig K. L., Shore R. E., Strax P., Pasternack B. S. A prospective study of endogenous estrogens and breast cancer in postmenopausal women.
J. Natl. Cancer Inst. (Bethesda)
,
87
:
190
-197,  
1995
.
11
Thomas H. V., Key T. J., Allen D. S., Moore J. W., Dowsett M., Fentiman I. S., Wang D. Y. A prospective study of endogenous serum hormone concentrations and breast cancer risk in post-menopausal women on the island of Guernsey.
Br. J. Cancer
,
76
:
401
-405,  
1997
.
12
The Endogenous Hormones and Breast Cancer Collaborative Group Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies.
J. Natl. Cancer Inst. (Bethesda)
,
94
:
606
-616,  
2002
.
13
Zhang Y., Kiel D. P., Kreger B. E., Cupples L. A., Ellison R. C., Dorgan J. F., Schatzkin A., Levy D., Felson D. T. Bone mass and the risk of breast cancer among postmenopausal women.
N. Engl. J. Med.
,
336
:
611
-617,  
1997
.
14
Cummings S. R., Duong T., Kenyon E., Cauley J. A., Whitehead M., Krueger K. A. Serum estradiol level and risk of breast cancer during treatment with raloxifene. The Multiple Outcomes of Raloxifene Evaluation (MORE) Trial.
J. Am. Med. Assoc.
,
287
:
216
-220,  
2002
.
15
Boyd N. F., Stone J., Martin L. J., Jong R., Fishell E., Yaffe M., Hammond G., Minkin S. The association of breast mitogens with mammographic densities.
Br. J. Cancer
,
87
:
876
-882,  
2002
.
16
Allred D. C., Mohsin S. K., Fuqua S. A. Histological and biological evolution of human premalignant breast disease.
Endocr. Relat. Cancer
,
8
:
47
-61,  
2001
.
17
Steeg P. S., Clare S. E., Lawrence J. A., Zhou Q. Molecular analysis of premalignant and carcinoma in situ lesions of the human breast.
Am. J. Pathol.
,
149
:
733
-738,  
1996
.
18
Dannenberg A. J., Altorki N. K., Boyle J. O., Dang C., Howe L. R., Weksler B. B., Subbaramaiah K. Cyclo-oxygenase 2: a pharmacological target for the prevention of cancer.
Lancet Oncol.
,
2
:
544
-551,  
2001
.
19
Early Breast Cancer Trialists’ Collaborative Group Tamoxifen for early breast cancer: an overview of the randomized trials.
Lancet
,
351
:
1451
-1467,  
1998
.
20
Fisher B., Costantino J., Wickerham L., Redmond C. K., Kavanah M., Cronin W. M., Vogel V., Robidoux A., Dimitrov N., Atkins J., Daly M., Wieand S., Tan-Chiu E., Ford L., Wolmark N. Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1.
J. Natl. Cancer Inst. (Bethesda)
,
90
:
1371
-1388,  
1998
.
21
Powles T., Eeles R., Ashley S., Easton D., Chang J., Dowsett M., Tidy A., Viggers J., Davey J. Interim analysis of the incidence of breast cancer in the Royal Marsden Hospital tamoxifen randomized chemoprevention trial.
Lancet
,
352
:
98
-101,  
1998
.
22
IBIS investigators (Writing committee: Cuzick, J., Forbes, J., Edwards, R., Baum, M., Cawthorn, S., Coates, A., Hamed, H., Howell, A., and Powles, T.) First results from the International Breast Cancer Intervention Study (IBIS-I): a randomized prevention trial.
Lancet
,
360
:
817
-824,  
2002
.
23
Veronesi U., Maisonneuve P., Costa A., Sacchini V., Maltoni C., Robertson C., Rotmensz N., Boyle P. Prevention of breast cancer with tamoxifen: preliminary findings from the Italian randomized trial among hysterectomised women. Italian Tamoxifen Prevention Study.
Lancet
,
352
:
93
-97,  
1998
.
24
Cuzick J., Powles T., Veronesi U., Forbes J., Edwards R., Ashley S., Boyle P. Overview of the main outcomes in breast-cancer prevention trials.
Lancet
,
361
:
296
-300,  
2003
.
25
Day R., Ganz P. A., Costantino J. P., Cronin W. M., Wickerham D. L., Fisher B. Health-related quality of life and tamoxifen in breast cancer prevention: a report from the National Surgical Adjuvant Breast and Bowel Project P-1 Study.
J. Clin. Oncol.
,
17
:
2659
-2669,  
1999
.
26
Tan-Chiu E., Costantino J., Wang J., Paik S., Butch C., Wickerham D. L., Wolmark N. The effect of tamoxifen on benign breast disease. Findings from the National Surgical Adjuvant Breast and Bowel Project (NSABP) Breast Cancer Prevention Trial (BCPT).
Breast Cancer Res. Treat.
,
69
:
210
2001
.
27
Goss P. E., Strasser K. Aromatase inhibitors in the treatment and prevention of breast cancer.
J. Clin. Oncol.
,
19
:
881
-894,  
2001
.
28
Smith I. E., Dowsett M. Aromatase inhibitors in breast cancer.
N. Engl. J. Med.
,
348
:
2431
-2442,  
2003
.
29
Nabholtz J. M., Buzdar A., Pollak M., the Arimidex Study Group Anastrozole is superior to tamoxifen as first-line therapy for advanced breast cancer in postmenopausal women: results of a North American multicenter randomized trial.
J. Clin. Oncol.
,
18
:
3758
-3767,  
2000
.
30
Bonneterre J., Buzdar A., Nabholtz J. M., Robertson J. F., Thurlimann B., von Euler M., Sahmoud T., Webster A., Steinberg M., the Arimidex Writing Committee, and the Investigators Committee Members Anastrozole is superior to tamoxifen as first-line therapy in hormone receptor positive advanced breast carcinoma.
Cancer (Phila.)
,
92
:
2247
-2258,  
2001
.
31
Mouridsen H., Gershanovich M., Sun Y., Perez-Carrion R., Boni C., Monnier A., Apffelstaedt J., Smith R., Sleeboom H. P., Janicke F., Pluzanska A., Dank M., Becquart D., Bapsy P. P., Salminen E., Snyder R., Lassus M., Verbeek J. A., Staffler B., Chaudri-Ross H. A., Dugan M. Superior efficacy of letrozole versus tamoxifen as first-line therapy for postmenopausal women with advanced breast cancer: results of a Phase III study of the International Letrozole Breast Cancer Group.
J. Clin. Oncol.
,
19
:
2596
-2606,  
2001
.
32
Baum M., Budzar A. U., Cuzick J., Forbes J., Houghton J. H., Klijn J. G., Sahmoud T., the ATAC Trialists’ Group Arimidex, tamoxifen alone or in combination. Anastrozole alone or in combination with tamoxifen versus tamoxifen alone for adjuvant treatment of postmenopausal women with early breast cancer: first results of the ATAC randomized trial.
Lancet
,
359
:
2131
-2139,  
2002
.
33
Fallowfield L., on behalf of the ATAC Trialists’ Group Assessing the quality of life (QOL) of postmenopausal (PM) women randomized into the ATAC (‘Arimidex’, tamoxifen, alone or in combination) adjuvant breast cancer (BC) trial.
Proc. Am. Soc. Clin. Oncol.
,
21
:
40a
(Abstract 159) 
2002
.
34
Goss P. E., Ingle J. N., Martino S., Robert N. J., Muss H. B., Piccart M. J., Castiglione M., Tu D., Shepherd L. E., Pritchard K. I., Livingston R. B., Davidson N. E., Norton L., Perez E. A., Abrams J. S., Therasse P., Palmer M. J., Pater J. L. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer.
N. Engl. J. Med.
,
349
:
1793
-1802,  
2003
.
35
Miller W. R., Dixon J. M. Antiaromatase agents: preclinical data and neoadjuvant therapy.
Clin. Breast Cancer
,
1(Suppl1)
:
S9
-S14,  
2000
.
36
Bulun S. E., Price T. M., Aitken J., Mahendroo M. S., Simpson E. R. A link between breast cancer and local estrogen biosynthesis suggested by quantification of breast adipose tissue aromatase cytochrome P450 transcripts using competitive polymerase chain reaction after reverse transcription.
J. Clin. Endocrinol. Metab.
,
77
:
1622
-1628,  
1993
.
37
Goss, P. E., Qi, S., Josse, R. G., Pritzker, K. P. H., Mendes, M., Hu, H., Waldman, S. D., and Grynpas, M. The effects of exemestane on bone and lipids in the ovariectomized rat. Bone, in press, 2003.
38
Goss P. E., Cheung A. M., Lowery C., Hu H., Qi S. Comparison of the effects of exemestane, 17-hydroexemestane and letrozole on bone and lipid metabolism in the ovariectomized rat.
Breast Cancer Res. Treat.
,
76(Suppl.1)
:
S107
2002
.
39
Goss P. E., Thomsen T., Banke-Bochita J., Lowery C., Asnis A. A randomized, placebo-controlled, explorative study to investigate the effect of low estrogen plasma levels on markers of bone turnover in healthy postmenopausal women during the 12-week treatment with exemestane or letrozole.
Breast Cancer Res. Treat.
,
76(Suppl.1)
:
S76
2002
.
40
Harper-Wynne C., Ross G., Sacks N., Dowsett M. A pilot prevention study of the aromatase inhibitor letrozole: effects on breast cell proliferation and bone/lipid indices in healthy postmenopausal women.
Breast Cancer Res. Treat.
,
69
:
225
2001
.
41
Harper-Wynne C., Ross G., Sacks N., Salter J., Nasiri N., Iqbal J., A’Hern R., Dowsett M. Effects of the aromatase inhibitor letrozole on normal breast epithelial cell proliferation and metabolic indices in postmenopausal women: a pilot study for breast cancer prevention.
Cancer Epidemiol. Biomark. Prev.
,
11
:
614
-621,  
2002
.
42
Heshmati H. M., Khosla S., Robins S. P., O’Fallon W. M., Melton L. J., III, Riggs B. L. Role of low levels of endogenous estrogen in regulation of bone resorption in late postmenopausal women.
J. Bone Miner. Res.
,
17
:
172
-178,  
2002
.
43
Winer E. P., Hudis C., Burstein H. J., Chlebowski R. T., Ingle J. N., Edge S. B., Mamounas E. P., Gralow J., Goldstein L. J., Pritchard K. I., Braun S., Cobleigh M. A., Langer A. S., Perotti J., Powles T. J., Whelan T. J., Browman G. P. American Society of Clinical Oncology technology assessment on the use of aromatase inhibitors as adjuvant therapy for women with hormone receptor-positive breast cancer: status report 2002.
J. Clin. Oncol.
,
20
:
3317
-3327,  
2002
.
44
Rossouw J. E., Anderson G. L., Prentice R. L., LaCroix A. Z., Kooperberg C., Stefanick M. L., Jackson R. D., Beresford S. A., Howard B. V., Johnson K. C., Kotchen J. M., Ockene J., Writing Group for the Women’s Health Initiative Investigators Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial.
J. Am. Med. Assoc.
,
288
:
321
-333,  
2002
.
45
Davies G., Martin L. A., Sacks N., Dowsett M. Cyclooxygenase-2 (COX-2), aromatase and breast cancer: a possible role for COX-2 inhibitors in breast cancer chemoprevention.
Ann. Oncol.
,
13
:
669
-678,  
2002
.
46
Half E., Tang X. M., Gwyn K., Sahin A., Wathen K., Sinicrope F. A. Cyclooxygenase-2 expression in human breast cancers and adjacent ductal carcinoma in situ.
Cancer Res.
,
62
:
1676
-1681,  
2002
.
47
Ragaz J., Brodie A., Jelovac D., Nielson T., Gilks B., Cheung M., Bajdik C., Spinelli J., Van de Rijn M. Expression of Cox2, aromatase, HER-2/neu and their correlation with the long-term outcome of 930 stage I-III breast cancer (BrCa) patients. Results from the British Columbia (BC) Tissue Micro-Array Project (BCTMAP).
Proc. Am. Soc. Clin. Oncol.
,
22
:
3
2003
.
48
Arun B., Zhang H., Mirza N. Q., Hortobagyi G. N., Hung M-C., Meric F. Growth inhibition of breast cancer cells by celecoxib.
Breast Cancer Res. Treat.
,
69
:
234
2001
.
49
van Landeghem A. A., Poortman J., Nabuurs M., Thijssen J. H. Endogenous concentration and subcellular distribution of estrogens in normal and malignant human breast tissue.
Cancer Res.
,
45
:
2900
-2906,  
1985
.
50
Brodie A., Jelovac D., Long B. J. Predictions from a preclinical model: studies of aromatase inhibitors and antiestrogens.
Clin. Cancer Res.
,
9(Suppl.)
:
455s
-459s,  
2003
.
51
Jelovac D., Long B. J., Sabnis G., Belosay A., Kataria R., Handratta V., Brodie A. M. H. Preclinical studies evaluating the anti-tumor effects of exemestane using the intratumoral aromatase postmenopausal breast cancer model.
Breast Cancer Res. Treat.
,
76(Suppl.1)
:
S77
2002
.
52
Chlebowski R. T., Col N., Winer E. P., Collyar D. E., Cummings S. R., Vogel V. G., III, Burstein H. J., Eisen A., Lipkus I., Pfister D. G., the American Society of Clinical Oncology Breast Cancer Technology Assessment Working Group American Society of Clinical Oncology technology assessment of pharmacologic interventions for breast cancer risk reduction including tamoxifen, raloxifene, and aromatase inhibition.
J. Clin. Oncol.
,
20
:
3328
-3343,  
2002
.