The oestrogen paradox: an hypothesis

Epidemiological and observational studies suggest that oestrogens, when used as hormonal therapy in post-menopausal women, can increase the risk of breast cancer if used long term. However, more recent data suggest that short-term use in sub-groups of post-menopausal women significantly decreases the risk of breast cancer. This beneficial effect is also observed when high-dose oestrogen is administered to post-menopausal women with breast cancer to cause tumour regression, a phenomenon which commonly occurs. We consider these divergent responses to oestrogen to represent a "paradox". Data from our own and other investigative groups suggest a hypothesis to explain this paradox. Deprivation of oestradiol in model systems causes cells to adapt and to undergo apoptosis in response to oestrogen. This occurs through the Fas/Fas ligand death receptor pathway and through alterations in apoptotic mechanisms mediated by mitochondria. This process of programmed cell death may explain the regression of established breast cancer with oestrogen administration and the diminution in the rate of new breast cancer diagnoses recently reported. Our hypothesis is based upon pathological data indicating the presence of a "reservoir" of undiagnosed breast cancer in the population of women who would be starting on oestrogens as menopausal hormonal therapy. The long-term increased risk of breast cancer may then reflect different mechanisms. Oestrogens can cause mutations through enhancement of the rate of cell division and concomitantly the error rate in DNA replication. In addition, oestrogens can be metabolised to directly genotoxic compounds. These carcinogenic processes take much longer, since a number of mutations must accumulate before resulting in breast cancer. These hypotheses regarding oestrogen-induced apoptosis in the short term and carcinogenesis in the long term now require rigorous verification but would serve to explain the "oestrogen paradox".


Introduction
As shown in Figure 1, a wide range of epidemiologic and observational data suggest that oestrogens are associated with the development of breast cancer [1,2]. With these data as a background, it was quite surprising that recently published data suggested that women taking postmenopausal hormone therapy (MHT) with oestrogen alone for 5 to 9 years unexpectedly experienced a decrease in the risk for breast cancer [3,4]. However, when taken for more than 20 years, the risk appeared to increase [5,6]. We call this the 'oestrogen paradox' to highlight the fact that short-term oestrogen use decreases the risk for breast cancer whereas long-term use increases it. A second component of the oestrogen paradox is that high-dose oestrogen therapy in postmenopausal women with breast cancer causes tumour regression, whereas the anti-oestrogen tamoxifen is equally effective in causing remissions in similar patient groups [7][8][9]. It is paradoxical then that both oestrogens and anti-oestrogens cause tumour regressions.

Short-term oestrogen use and breast cancer risk
The initial publication of the Women's Health Initiative (WHI) [3] reported a 23% decrease in invasive breast cancer incidence in patients taking oestrogen alone compared with placebo, a finding which narrowly fell short of statistical significance (hazard ratio [HR] 0.77, 95% confidence interval [CI] 0.59 to 1.01). A recent exploratory analysis of updated data from this study examined subgroups to determine whether oestrogens might reduce the incidence of breast cancer significantly in women falling into certain categories [4]. Notably, this analysis reported a statistically significant 33% reduction in invasive breast cancer incidence in patients who strictly adhered to their oestrogen therapy (HR 0.67, 95% CI 0.47 to 0.97). In addition, a 31% lower incidence of localized breast cancer (HR 0.69, 95% CI 0.51 to 0.95) and a 29% reduction in ductal cancers (HR 0.71, 95% CI 0.52 to 0.99) were reported in oestrogen users. The decreases in breast cancer risk were limited to women who had not previously used MHT [4]. In a concurrent report from the Nurses Health Study [2], a significant 26% decrease in risk for breast cancer was observed in obese women, and a nonsignificant 10% decrease in all study participants, taking oestrogen alone for 5 to 9 years. Other observational studies reported a reduction in risk with oestrogen alone but of lesser magnitude and not statistically significant. For example, Schairer and colleagues [5] reported a 7% reduction in breast cancer risk at 6 years in women receiving oestrogen alone, and Lyytinen and coworkers [10] identified a similar 7% reduction. These combined results, although not conclusive, are highly suggestive of a beneficial effect of oestrogen in reducing breast cancer risk. However, this conclusion must be considered provisional until rigorous confirmation in additional studies is reported.

Long-term oestrogen use and breast cancer risk
What are the data regarding use of oestrogen alone for more than 20 years? The Nurses Health Study [2] also evaluated women using oestrogen alone for more than 20 years and found a statistically significant 41% increase in breast cancer risk in women 50 years of age or older, and a 77% increase in the subset of lean women. Earlier studies by Magnusson [11] and Schairer [5] and their colleagues also identified significantly increased breast cancer risks in women taking oestrogen alone for more than 10 years (odds ratio 2.7) and 16 years (relative risk 1.6), respectively. The Million Women Study [6] also reported a linear increase in breast cancer risk over time in women receiving MHT with oestrogen alone over a period of 10 years. In contrast to the other studies reported, however, the Million Women Study found a nonsignificant increased risk for breast cancer, even in women receiving this therapy for less than 5 years.

High-dose oestrogens as breast cancer treatment
A second component of the oestrogen paradox is that women with hormone-dependent breast cancer respond to high-dose oestrogens with objective tumour regressions. This form of therapy was the mainstay of hormonal treatment of breast cancer from the late 1940s until the early 1980s [7][8][9]. When compared in randomized trials with tamoxifen, highdose oestrogens were equally efficacious [7] and in one study they were associated with significantly enhanced survival [8] compared with an anti-oestrogen. Extensive studies demonstrated that only specific subgroups of women respond to high-dose oestrogen [9,12]. Premenopausal women and those less than 1 year postmenopausal do not respond at all. Women who had undergone menopause many years earlier frequently experienced objective tumour regressions; the longer the duration of the period after cessation of menses, the greater the response rate. Only oestrogen receptor (ER)-positive tumours regress in women receiving high-dose oestrogens [12].

Possible mechanisms to explain the oestrogen paradox
Our preclinical data demonstrate that long-term deprivation of oestradiol causes this sex steroid to trigger cell death through apoptosis (Figure 2a), whereas wild-type cells with a normal oestrogen milieu exhibit reduced apoptosis (Figure 2b) [13][14][15][16][17][18][19][20][21]. The postmenopausal women receiving MHT with oestrogen alone may be considered to be in a state of longterm oestradiol deprivation. Extensive review of autopsy studies provides strong evidence that there is a reservoir of undiagnosed breast cancer in postmenopausal women (Table 1) [22,23]. The short-term reduction in breast cancer in the patients with undiagnosed occult breast tumours may be due to oestrogen-induced apoptosis of tumour cells. Similarly, the effect of oestrogen in inducing tumour regressions in patients with known breast cancer may reflect a similar phenomenon. We suggest that the increased risk for breast cancer results from long-term use of oestrogens alone because the risk from MHT may occur via different mechanisms [24,25]: the genotoxic effects of oestradiol metabolites and the ER-mediated proliferative effects of oestradiol. The following sections of this treatise will review the evidence for each of these statements.

Occult pre-existing breast cancers in women
Over the past three decades at least eight studies have assessed the frequency of occult malignant disease, primarily ductal carcinoma in situ (DCIS), found at autopsy in women with no history of breast cancer [22] (Table 1). The frequency of occult DCIS varied considerably among these studies (range 0% to 15%), most likely reflecting methodological differences. Variation aside, approximately 5% of the 1,052 combined cases from these studies included occult DCIS and 1% occult invasive breast cancers [22]. Based on these findings, it is reasonable to assume that 5% to 10% of the women entering the WHI and Nurse's Health Study had occult breast cancer when they were initially enrolled.

Evidence for oestradiol-induced apoptosis
Recent in vitro studies from our laboratory showed that hormone-dependent breast cancer cells deprived of oestrogen in the long term undergo adaptive changes that cause oestrogen to paradoxically stimulate apoptosis [13][14][15] (Figure 2a). Whereas wild-type MCF7 cells respond to oestradiol with a reduction in apoptosis, those deprived of oestrogen in the long term exhibit an increase in programmed cell death. Similarly, Jordan and collaborators [16][17][18][19][20][21]  demonstrated that long-term tamoxifen exposure also results in adaptation and development of oestrogen-induced apoptosis. Apoptotic mechanisms in adapted cells involve upregulation of death receptor as well as mitochondrial pathways. Specific molecular events include activation of the Fas death receptor/Fas ligand complex, the release of cytochrome C from the mitochondria, alterations in Bcl-2, and downregulation of the anti-apoptotic factor nuclear factor-κ [14,15,18].

Long-term oestradiol deprivation in the Women's Health Initiative and the Nurse's Health Study
At the time of enrolment, participants in the WHI trial were 63 years old on average and menopausal for more than 10 years [3]. Plasma oestradiol levels fall precipitously at menopause from 50 to 600 pg/ml to levels of 5 to 10 pg/ml. Even though breast tissue levels might not precisely reflect plasma concentrations, one would still expect substantial reduction in breast tissue levels and adaptation to this reduction. If our hypothesis were correct, then exposure to oestrogen therapy as MHT would induce apoptosis and shrink or even eradicate the occult tumours, which would reduce the detection of a cancer by mammography or palpation over the next several years. This scenario could explain the reduction in breast cancers diagnosed in the WHI and Nurses Health Study in women receiving oestrogen alone as MHT for 5 to 9 years [2,4]. This hypothesis would also explain why women who had received MHT before entering the WHI study did not experience a reduction in breast cancer risk [4].

Long-term exposure to oestradiol
Why would oestrogen increase the risk for breast cancer when it is given for more than 20 years? The commonly accepted explanation for the carcinogenic effect of oestrogen is that this sex steroid stimulates breast cancer proliferation genes, increases the rate of breast cell divisions, and thereby enhances the chances for development of mutations [25]. An additional and more controversial mechanism suggests that metabolites of oestradiol are directly genotoxic [24,25] ( Figure 3). Recent studies demonstrate that oestradiol is converted to 4-OH-oestradiol in human breast tissue via the cytochrome p450 1B1 enzyme, and it is then oxidized to quinone metabolites. These metabolites are highly reactive and covalently bind to adenine and guanine on DNA, resulting in depurination, error-prone DNA repair, and point mutations [24]. Other recent studies have shown that 4-OH-oestradiol is directly mutagenic in cellular mutagenesis assays [26][27][28][29]. In addition, 4-OH-oestradiol can transform ER-negative benign breast epithelial cells into serially transplantable  carcinomas in immune deficient mice [28]. Finally, an ER knockout model of breast cancer forms tumours in response to increasing doses of exogenous oestradiol in previously castrated animals [24,30]. These combined observations suggest that directly genotoxic as well as ER-mediated mechanisms may be responsible for the long-term carcinogenic effects of oestradiol [24]. In time, the pro-carcinogenic effects of oestradiol would outweigh the pro-apoptotic effects.

Conclusion
A variety of data are congruent with our 'oestrogen paradox' hypothesis; however, additional confirmatory studies are needed to prove this contention. Specifically needed are more comprehensive autopsy studies to determine precisely the magnitude of the reservoir of occult breast cancers and their precursor lesions. The ability of highly sensitive imaging strategies, such as digital mammography and magnetic resonance imaging, should be evaluated for their abilities to detect occult breast cancers in women initiating MHT. Direct demonstration of oestrogen-induced apoptosis in occult breast cancers in women will also be critical.