Hormone therapy and breast cancer

Introduction

The role of hormones in the development of cancers has been a point of discussion and dispute for a number of years. Although some causal relationships have been easily proven, such as the role of estrogen in the development of endometrial cancers, in other cancers the relationship has been more difficult to define. Our understanding of hormonal physiology continues to define variable effects on different tissues, which may lead to specific carcinogenic properties. For example, we know that progestins are protective against the proliferative effect of estrogen in the uterus, but in the breast have a more complex influence and may both decrease breast cancer risk by opposing estrogenic stimulation of the breast[1] and increase risk by increasing breast mitotic rates.[2] Recent evidence suggests that progesterone does not down-regulate estrogen and progesterone receptors in the breast,[3] and may catalyze the conversion of the less potent estrone to the more potent estradiol.[4] These activities may vary according to both the age and time of menstrual cycle of a woman, adding further complexity to a poorly understood physiology.

When reviewing hormones, two issues must be considered. One is the risk that hormones have in the development of cancer in persons with no history of malignancy. The second issue is the use of exogenous hormones in women with a previous history of cancer.

Hormones as a cause of breast cancer

Hormones obviously have a role in the epidemiology of breast cancer, and a comprehensive discussion is beyond the scope of this article. Issues of family history, reproductive history, age of menarche, menstrual cycle characteristics, lactation, age of menopause, and obesity as it relates to estrogen levels are all well established factors.[5] More recently studies of various ethnic groups, particularly in areas of the world with lower rates of breast cancer, have further emphasized the different hormone levels and menstrual cycle patterns and drawn attention to the prolonged, cyclic stimulation of breast tissue in women in our society. Models of breast cancer risk taking into account parity and age of menarche are available and closely match the observed rates. Among postmenopausal women, the most consistent finding is a positive relationship between total estradiol and the risk of breast cancer. Breast cancer patients have a mean estradiol level 15% higher than that of healthy controls.[6] As well, higher testosterone and dehydroepiandrosterone (DHEA) levels have been positively associated with breast cancer in most studies of postmenopausal women, and in some studies show only a modest effect when estrogen levels are considered, suggesting an indirect effect.

Prospective studies of hormone levels and exogenous hormone use have often been limited by a lack of consistent findings, possibly due in part to differing methodologies, limited sampling times of hormone levels not reflecting the woman’s full cycle, and poor patient recall of hormonal use. More than three dozen epidemiologic studies have been done over the past 20 years to try to establish safety and risks, and have been summarized in six meta-analyses and a large pooled analysis.[7-11] All the meta-analyses have concluded that women who have ever used HRT have a small increased relative risk ranging from 1.01 to 1.14. The “ever used” category, however, fails to distinguish between short and long duration and recent and past users. In the meta-analyses, significant increases in risk were seen with more than 5 years of use, but were not confirmed in several large case-controlled studies. The updated Nurses’ Health Study, which included 1935 breast cancer cases, reported that an excess risk of breast cancer was limited to women with current or very recent use of hormones. The risk increased with increasing duration of use and was statistically significant among current users of 5 or more years’ duration compared with those who had never used postmenopausal hormones. The relative risk for those with 10 or more years of use was 1.47 (95% confidence interval 1.22–1.76).[12,13] The recent meta-analysis by the Collaborative Group confirmed that the excess risk of breast cancer was more pronounced in women with current or recent use of exogenous hormones,[14] but failed to show an impact on survival. This may suggest that the cancers were diagnosed early in women closely monitored on HRT, that the cancers that developed on HRT were less aggressive and more curable, or that there were not enough deaths at the time of analysis to make an impact. The type of estrogen has not been consistently shown to be a factor.

The combination of progesterone and estrogen is the most commonly prescribed hormone replacement prescription. Although the recent study reported in JAMA received widespread publicity, the results were not significantly different than previous reports in the literature.[15] The relative risk of developing breast cancer increased by 0.01 for each year of estrogen-only use and 0.08 with each year of combined estrogen-progestin use for current and recent users. There was an increased risk for lean women, that is, women with a body mass index of 24.4 kg/m² or less. This study once again highlighted the different actions of the hormones on different organs and confirmed that what is safe for the uterus may carry risk for breast tissue. At this time, however, in the woman with an intact uterus who is receiving HRT combination therapy should remain standard.

Thus, although studies show a modest increase in the relative risk of developing breast cancer for women on HRT or women with elevated estradiol levels, these studies are not balancing the benefits of HRT with the risk, but simply looking at the malignancy endpoint. Retrospective reviews have attempted to balance these effects by reviewing cause of death in women and years of life lost or gained by HRT, and are probably still accurate, although disease incidence may change over time[16-18] (Table 1). Prevention of menopausal symptoms and conditions should be tailored to each woman, and attempted by a combination of lifestyle measures including diet and exercise, hormonal replacement for a limited duration in appropriate women, and newer medications, such as the bisphosphonates, with beneficial effects. This subject is more fully discussed in Dr Montemuro’s paper in the first part of this theme issue. Simple avoidance of HRT due to fears of cancer is not warranted, but a woman’s concerns about increasing her cancer risk are appropriate and should be approached with information and options.

Hormone therapy in patients with a history of breast cancer

The traditional dogma has been that any use of hormones in patients with a personal history is contraindicated. This is based on the epidemiology of primary breast cancer, which reports an increased incidence of breast cancer in postmenopausal women exposed to estrogen replacement therapy.[19] As well, this 1997 study has confirmed that HRT using a combination of progesterone and estrogen is also associated with an increased risk of developing breast cancer.

There is no direct evidence that estrogen causes recurrence in women with a previous history of cancer, but there is evidence of a decreased recurrence rate when women with a history of breast cancer are treated with tamoxifen or oophorectomy. Concerns about doing harm and contributing to incurable breast cancer recurrences have limited the use of estrogen replacement in these women. However, we are seeing an increased number of young women with early breast cancer and a relatively good prognosis who are experiencing early menopause after adjuvant chemotherapy. Although not universal with chemotherapy, increasing age, longer-duration chemotherapy protocols, and more intensive dosing all contribute to ovarian dysfunction. As well, through screening mammography, we are identifying large numbers of older women with curable early breast cancers, some of whom are interested in continuing their hormone replacement therapy.

There is currently no reliable information on the risks of hormone replacement therapy to women after a diagnosis of breast cancer. There are a number of series reported in the literature that are not helpful because the studies are too small and have very short follow-up.[20-23] There are plans for large, randomized studies of women with a history of breast cancer to assess the risks and benefits of hormone replacement therapy, but this data will not be available for many years.

Without evidence of safety, the general policy of the BC Cancer Agency Breast Tumour Group has been to do no harm and therefore to not risk prescribing HRT to women with a previous history of any breast malignancy if there are other therapeutic options that can result in similar outcomes. Although there is good evidence that HRT can improve bone health and some contradictory evidence regarding heart health, there are often other therapies and lifestyle changes that can also be of benefit and should be considered as the first options. In situations where there is a greater health risk from the non-breast cancer disease and there are no other options, the woman should be clearly informed of the situation and an individual decision to use HRT may be appropriate. Similarly, if there are symptoms that interfere significantly with a woman’s quality of life (i.e., troublesome hot flushes, vaginal dryness, vaginal atrophy, dyspareunia, urinary incontinence, sleep disturbances, depression) and other therapeutic options have failed, HRT should be considered. For vaginal complaints, topical estrogens may be considered. With estrogen cream there may be some systemic absorption. An estradiol vaginal ring (Estring) causes a transient increase in estrogen levels on day 1 but subsequently there is no measurable systemic effect. In all situations the woman must be fully informed about the risks, benefits, and areas where we lack clear information. The lowest dose of hormone replacement that is effective should be used for a limited duration.

There is some confusion about many of the new selective estrogen receptor modifiers (SERMS) and their role in HRT and breast cancer. Tamoxifen (acting as an anti-estrogen) has been shown to be of benefit in decreasing relapse and new cancers in women with a history of in situ and invasive breast cancer.[24] Based on an individual risk assessment, tamoxifen may be prescribed for women for a duration of 5 years. Similarly there is evidence of tamoxifen being of benefit for the prevention of breast cancer in high-risk women, and it may be appropriate to prescribe tamoxifen for selected, informed high-risk women.[25] High risk was defined in the National Surgical Adjuvant Breast and Bowel Project prevention study as a greater than 1.7% risk of developing breast cancer over 5 years, but this threshold has been modified by most experts.[26] Tamoxifen has rare side effects including venous thromboembolism, endometrial cancer, and cataracts, so the benefits must outweigh the risks (see Table 2 for a summary of the provincial guidelines). Models[27,28] have been used to assess an individual woman’s risk and have been adapted by the Breast Tumour Group.[29] The Gail model has been used to calculate 5-year and lifetime risks of breast cancer (Table 3) and is available as a simple program from the National Cancer Institute at www.cancernet.nci.nih.gov/h_detect.html or the BC Cancer Agency library.

Raloxifene (Evista) is currently being studied in a large randomized study comparing it to tamoxifen for prevention of breast cancer in postmenopausal women. Information on this study is available at (604) 822-7997. It is not approved or proven for use in women with a history of breast cancer, nor is it approved for prevention. The osteoporosis study[30] suggested a 76% decrease in the rate of breast cancer in a group of women who were not at high risk. However, this study was in older women, had very small numbers, was not planned to have breast cancer as a primary endpoint, and needs confirmation. Both raloxifene and tamoxifen can cause hot flushes and dyspareunia, and are not appropriate to treat these menopausal symptoms. Both drugs are effective in improving heart and bone outcomes, and raloxifene is licensed for use in osteoporosis. We are all waiting for the perfect SERM that will prevent osteoporosis, cardiac disease, Alzheimer’s, and breast cancer, relieve hot flushes, vaginal complaints, and depression, and not increase the risk of thromboembolic events or uterine cancer. Unfortunately that SERM is not yet available.

There is continued controversy surrounding the use of progestins as single agents for postmenopausal symptoms and health. A number of research studies are ongoing to clarify their safety and risks in breast cancer. In contrast to the well-established role progestins have in decreasing uterine cancer when used in combination with estrogens, they are associated with an increased risk of breast cancer when used in combination with HRT. As outlined above, recent studies have shown an increased risk of breast cancer when progesterone-estrogen combinations were used in women with no previous history of breast cancer,[31] and one can presume a similar risk for women with a previous diagnosis. This, however, should not preclude their use in an informed, symptomatic woman with an intact uterus.

Research is ongoing on phytoestrogens for symptom control after a diagnosis of breast cancer. Studies, including a randomized, double-blind study from the BC Cancer Agency, have suggested that soy protein may not be helpful in relieving hot flushes. Other studies are pending. There is conflicting evidence on the potential actions of phytoestrogens, and it appears in some studies that their estrogen agonist or antagonist activity may be related to dose, but again there is currently no hard evidence that can translate into clear patient recommendations. Many other treatments are also under investigation.

Hormones in women with a BRCA genetic mutation

Although less than 10% of the breast cancers diagnosed at this time in British Columbia are in women with one of the defined heritable genetic mutations, there is interest in the role of hormones in this group of women. The role of HRT in women with a known genetic mutation in BRCA1 or BRCA2 causing an increased risk of breast or ovarian cancer is unknown. There is evidence that oral contraceptives are of value in decreasing the incidence of ovarian cancer.[32] It is not clear if the incidence of breast cancer is affected by either HRT or oral contraceptives. There is one study suggesting that early and multiple pregnancies are not protective in these women, but more research is needed. These issues are of particular importance to women with an identified mutation who have not yet developed a malignancy. Many centres advocate oophorectomy at either age 35 or once a woman with an identified mutation no longer wants the ability to become pregnant because of the high lifetime risk of ovarian cancer and our limited ability to screen for it. Surgery is followed by hormone replacement therapy, but the safety for a women who has intact breasts is not yet established. Again, a balance of health issues, symptoms, and risks must be identified for each woman, recognizing that we have not yet established the role of hormones in the development of cancers in women with BRCA mutations.

Summary

The role of HRT in women with a history of breast cancer is unknown. If there are options for other therapies, they should be used. In individual informed cases HRT may be appropriate for some women. Research may give us new insights in the next few years. If there are patients that need further assessment, please consult their oncologists. If HRT is used, it may be advisable to prescribe the lowest dose of estrogen to relieve symptoms and to monitor the patient carefully. Consider short-term use until long-term data is available. Although SERMS may be of value in preventing chronic complications of menopause, their use for the acute symptoms is limited and may exacerbate them. Until more effective and proven therapies are available, it is important that we stay informed, educate our patients, and try to facilitate the rational and informed use of HRT.

Karen A. Gelmon, MD, FRCPC and Charmaine Kim-Sing, MD, FRCPC

Dr Gelmon is a clinical professor in the Department of Medicine at the University of British Columbia and head of the Provincial Breast Tumour Group at the British Columbia Cancer Agency. Dr Kim-Sing is a clinical assistant professor in the Department of Medicine at UBC and the medical leader of the Hereditary Cancer Program at the British Columbia Cancer Agency.