Intrabreast estrogen levels and the development of breast cancer

Aromatase activity probably occurs within the breast. For reasons that are unclear to me, in postmenopausal women the breast increases its estrogen production. It might just be because the estrogen levels fall; however, there is almost parity in intrabreast estrogen concentrations in pre- and postmenopausal women, which is extraordinary when you think of the reduction in plasma estrogen levels in postmenopausal women.

I believe women at risk for breast cancer have over-estrogenized breasts. It might be because there’s receptor quirkiness or some coactivator milieu — I don’t know why, and I’m not saying it’s the only mechanism of breast cancer development, but I truly believe that breast cancer is usually caused by overestrogenecity of the breast. I think it’s like a subtle and slow poison that builds year after year.

There is marked ratio of intrabreast to peripheral levels of estrogen. It has been suggested that a very low dose of an aromatase inhibitor may shut off the intrabreast aromatase production sufficiently to tone down the estrogen level. Even a slight reduction in estrogen could translate into a profound reduction of risk. We may not need to obliterate estrogen; rather we may just need to tone it down more specifically in the breasts.

Modulating the aromatase gene to reduce estrogen production in the breast

The HER2, COX-2 and aromatase are on a hierarchical pathway, and they actually drive each other. In my opinion, you could shut the pathway down at any one of those levels.

The COX-2 pathway is of specific interest to us. It’s induced by the presence of ductal carcinoma in situ and by invasive cancer. COX-2, through prostaglandin E-2 and modulated through cyclic AMP, upregulates the aromatase gene and causes estrogen production. In the preinvasive lesion, one thing that could be exploited is that as the cells start progressing to DCIS, the COX-2 pathway starts to increase breast production of estrogen. This could be toned down with COX-2 inhibition. I think celecoxib alone could be an intrabreast cancer drug, and indeed the epidemiologic data supports that.

In the hormone-dependent rat model, celecoxib acts against estrogen receptorpositive breast cancer. In cultured cells, it acts against estrogen receptor-negative breast cancer cells, causing a dose-dependent reduction in proliferation. In the rat mammary model, it has synergy with exemestane. Celecoxib has the potential to help exemestane knock out ER-positive lesions and to impact ERnegative lesions independently by blocking the COX-2 pathway.

A Phase III chemoprevention trial of exemestane and celecoxib

The NCIC of Canada will launch a worldwide prevention trial comparing placebo versus exemestane versus exemestane plus celecoxib. One rationale is that we think there is a higher proportion of hormone-dependent lesions in the preinvasive disease setting than in the invasive disease setting, so we believe antihormone therapy will have its greatest impact in prevention.

There were several reasons for incorporating a placebo arm, including that it’s easier to show true efficacy and toxicity of a compound against a placebo and that the sample size of the study is much smaller. In addition, the meta-analysis of tamoxifen, particularly in elderly women, suggests no net health benefit. Dr Jack Cuzick has applied those data to the ASCO Technology Assessment, and the expert panel recommended a placebo for future breast cancer prevention trials.

Side effects of aromatase inhibitors: Implications for prevention

Considerably fewer vasomotor symptoms and problems with weight gain are associated with aromatase inhibitors than with tamoxifen. While these are anecdotal observations, I have seen these differences in my own practice so often that I’m fairly certain they will prove to be true.

Perfectly healthy women considering prevention have a different level of motivation and tolerance of side effects than breast cancer patients who have been thrust into menopause by chemotherapy. The aromatase inhibitors are very well-tolerated and very safe, and I think healthy women with even the slightest motivation to reduce their breast cancer risk will find them acceptable.

Impact of the ATAC data on clinical practice

I was taken aback by the ASCO Technology Assessment. I agree with their points, but I think the onus on the regulators wasn’t to see if anastrozole was better than tamoxifen, but only to see if it was worse. It’s almost inconceivable that it could turn out to be worse.

Now, with 13 more months of follow-up, I believe the data is going to change people’s viewpoints. My personal take on the ATAC presentation in 2001 was that we should switch from tamoxifen to anastrozole for adjuvant therapy. The curves were convincing, and the trend is likely to increase with time.

When selecting an aromatase inhibitor, clinically, it makes sense to use anastrozole because we have the data to support it. From a research perspective, it’s probable that other compounds will yield the same or even better results. Medical-legally, it would be difficult to defend the choice of another aromatase inhibitor for which there is no data over anastrozole in the clinical setting.

Hormonal therapy after failure on adjuvant anastrozole

Selection of a hormonal therapy after a patient relapses on anastrozole is a problem. Tamoxifen or fulvestrant could be highly effective, but if the MAP kinase pathway is overdriven from the aromatase inhibition, tamoxifen might act more as an agonist, and fulvestrant might be a better choice. To my knowledge, in terms of ATAC or other patients who have relapsed on an adjuvant aromatase inhibitor, there haven’t been any data presented yet addressing this issue.

Benefits of bisphosphonate therapy

The impact of adjuvant aromatase inhibitors on bone will be offset by bisphosphonate therapy. Hopefully, in the case of exemestane, it won’t even be an issue. Bisphosphonates cause osteoclast apoptosis and inhibit osteoclast activity, so they should be able to counteract the estrogen depletion effects of aromatase inhibitors. Estrogen deprivation increases bone re-absorption, whereas bisphosphonates decrease it. Some data suggest bisphosphonates have an antimetastatic effect, so they may be more than just a salvage therapy for the aromatase inhibitors; they may also be an anticancer therapy.

Effect of estrogen levels on women’s health

I am very interested in how closely a woman’s health is related to a small range of estrogen. The slope of the postmenopausal estrogen range is a very small curve and is tightly related to breast cancer risk, osteoporosis and probably cardiovascular risk.

My personal feeling is that healthy postmenopausal women without breast cancer need their estrogen level “customized.” We can tone it up with HRT, but now we have a subtle way of toning it down. We have tried the two extremes — blockbuster ablation or massive replacement — but we haven’t tried zoning in on a middle range. I’m 100 percent convinced that in the next 10 to 15 years we will develop an understanding of women’s estrogen levels and their impact on health.

Select publications

Home · Search