| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Toronto-Sunnybrook Regional Cancer Centre and The University of Toronto, Toronto, Ontario, Canada
Correspondence: Kathleen I. Pritchard, M.D., Toronto-Sunnybrook Regional Cancer Centre, 2075 Bayview Avenue, Toronto, Ontario M4V 1H6 Canada. Telephone: 416-480-4616; Fax: 416-480-6002; e-mail: kathy.pritchard{at}tsrcc.on.ca
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
Key Words. Breast cancer • Hormone replacement therapy • Estrogen • Progesterone
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
|
MENOPAUSAL PHYSIOLOGY |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
|
ERT/HRT IN TREATMENT OF MENOPAUSAL SYMPTOMS |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
|
LONG-TERM POSITIVE EFFECTS OF ERT/HRT |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
Cardiovascular
The role of estrogen in CVD is less clear because most studies are observational rather than interventional. It is known that estrogen with or without progesterone increases high density lipoprotein (HDL) cholesterol and decreases low density lipoprotein (LDL) cholesterol [8, 9], but this is believed to represent only part of its action. Estrogen also has other effects on the cardiovascular system including direct action on vessel walls, specific effects on the myocardium [10] and effects on platelets and other coagulation factors. One randomized trial has shown an effect of estrogen alone or with a variety of progestationals in increasing HDL and reducing LDL in women over a 3-year period compared to a placebo control. This trial was not powered to examine cardiovascular end points, however [11].
Investigators have commonly inferred reductions in CVD events of the order of 40% from case-control and cohort study results [12] for HRT users compared to nonusers. Because these are observational studies, however, patient and physician selection factors may result in more healthy women being the women who receive ERT. A recent meta-analysis of coronary heart disease (CHD) end points in 22 available randomized trials which were primarily designed to study other outcomes in 4,124 postmenopausal women found no effect of HRT on CHD events (odds radio = 1.39; 95% confidence intervals = 0.48 to 3.95) [13]. The only published randomized trial of ERT/HRT with CHD events as a primary end point, the Heart and Estrogen/Progesterone Replacement Study (HERS), a randomized trial of estrogen plus progestin for secondary prevention of CHD in postmenopausal women, showed that women started on such HRT shortly after a cardiac event were more likely to suffer a second cardiac event over the next year. As these patients were followed further, however, those randomized to receive HRT were less likely to suffer a second cardiac event in years 4 and 5 of follow-up, so that there was overall no significant difference in cardiac morbidity in those randomized to HRT or placebo [14]. Since this randomized trial showed early results that are opposite to those expected based on observational data, there has been some reexamination of the assumption that the results of observational studies in this area will be duplicated in randomized trials. The Women's Health Initiative (WHI) Study comparing ERT/HRT to placebo in over 25,000 postmenopausal women is only part way through its accrual [15]. This trial will provide the first randomized evidence of ERT/HRT influence on primary CVD end points and overall mortality.
Alzheimer's Disease and Cognitive Function
Several observational studies have suggested a relationship between ERT/HRT and improved cognitive function or reduced risk of Alzheimer's disease [16]. These studies may also be subject to selection bias. However, the recently published Alzheimer's Disease Cooperative Study of Mulnard and coworkers, which randomized 97 women with mild to moderate Alzheimer's disease to low-dose (0.625 mg) or high-dose (1.25 mg) estrogen or placebo daily clearly showed no such effect. After 1 year the average score on the 7-point Clinical Global Impression of Change scale for women receiving estrogen was 5.1 compared with 5.0 for women taking placebo. There was no significant difference between the groups in measures of mood, memory, attention, language skills, motor function, or activities of daily living [17]. The authors feel that although these results are clearly negative, they relate to only the specific group of women already 75 years of age and having Alzheimer's. It is still possible that estrogen can improve cognition in women in mid-life or in older women without Alzheimer's disease. Currently the influence of ERT/HRT on memory and mental function is undergoing prospective evaluation in a randomized trial in postmenopausal women aged 65-79 in the WHI Memory Study [15]. Once again however, results will not be available for 5 or 6 years.
Colon Cancer
It has also been demonstrated, in a series of observational studies, that the risk of colon cancer is considerably lower in association with the use of ERT/HRT. A relative risk of as low as 0.5 in these studies would seem to suggest a real association [18]. Colon cancer incidence will be one of the outcomes measured in the WHI study.
|
ALTERNATIVES |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
Estrogen Deficiency Symptoms
Estrogen deficiency symptoms can be managed by a variety of alternatives [19, 20]. KY Jelly and Replens can significantly reduce vaginal dryness and local menopausal symptoms [21, 22]. Other approaches for persistant local symptoms include vaginal estrogen creams and Estring [23], which are, however, known to be associated with vaginal absorption of estrogen to levels which may, in some cases, be comparable to those achieved with oral use [24]. Estring tends to provide more consistent local effects with lower systemic absorption than the use of creams. Hot flashes can be treated with a variety of nonhormonal therapies.
Because of the perceived unacceptability of ERT/HRT in women with a previous diagnosis of breast cancer, the Mayo Clinic and North Central Cancer Treatment Group (NCCTG) have conducted a series of clinical trials involving over 650 cancer survivors to look at various methods for alleviating hot flashes. In this series of clinical trials [25-28], the effect of a placebo on hot flashes is well illustrated, demonstrating that it causes a relatively consistent 20%-25% reduction in hot flashes over a 4-week period. It is not clear how much of this is actual placebo effect versus the natural history of hot flashes to diminish with time. This placebo effect needs to be taken into consideration, however, when evaluating new agents and understanding anecdotal experiences.
The NCCTG recently completed a placebo-controlled trial looking at a soy phytoestrogen preparation. The magnitude of interest in this compound for this symptom is illustrated by noting that 180 patients were entered on this clinical trial over a 2-month time period. Unfortunately there was no suggestion that soy protein significantly reduced the severity or frequency of hot flashes in comparison to placebo. At study completion, patients preferred the soy product 33% of the time, the placebo 31% of the time, and neither substance 31% of the time [29].
The NCCTG also recently completed a placebo-controlled trial of vitamin E 800 I.U. per day [27]. This clinical trial did demonstrate that vitamin E was able to statistically significantly decrease hot flashes over placebo. However, this hot flash reduction amounted to one hot flash per person per day. The vitamin E was well tolerated in this clinical trial.
Another NCCTG placebo-controlled hot flash trial demonstrated that clonidine could reduce hot flashes by approximately 15% more than placebo [25]. Nonetheless, in this clinical trial, clonidine was associated with statistically significantly more toxicity and patients did not prefer it over placebo at study end.
The further NCCTG hot flash trial evaluated a low-dose of megestrol acetate compared to a placebo [26]. This trial demonstrated a hot flash reduction of approximately 80% with megestrol acetate. The therapy was well tolerated in this short-term double-blind, crossover clinical trial and women preferred megestrol acetate significantly more than placebo. A subsequent investigation suggested that megestrol acetate continues to control hot flashes for up to 3 years of therapy [30]. Most women who continued to use megestrol acetate were able to utilize a dose of 
20 mg per day with effective control of hot flashes.
A pilot trial also conducted at the Mayo Clinic suggested that a very low dose of the relatively new antidepressant, venlafaxine (Effexor), was able to decrease hot flashes by approximately 50% [28]. This low dose of venlafaxine appeared to be relatively well tolerated overall. Anecdotal information has suggested that other selective serotonin reuptake inhibitors (SSRIs) also can decrease hot flashes, leading to a number of other ongoing clinical trials. A placebo-controlled dose-finding clinical trial is in development in the NCCTG to more definitively determine the efficacy and potential toxicity of venlafaxine for hot flashes in breast cancer survivors.
Thus, depending on hot flash severity, patient preferences, patient willingness to undertake theoretical risk, and physician prejudices, there are a number of options. Vitamin E can be utilized, as it statistically significantly decreases hot flashes. This medication is inexpensive, nontoxic, and readily available. It would allow a patient to get the well-described placebo effect and maybe a bit more. Nonetheless, vitamin E has limited efficacy. Clonidine is an option that some physicians utilize given the information that it can decrease hot flashes. The demonstrated increase in toxicity, however, needs to be factored into the decision as to whether to utilize this in clinical practice. Given the promising preliminary information described above, low doses of venlafaxine are reasonable to try, pending results from randomized placebo-controlled clinical trials. A daily dose of 37.5 mg in a sustained release preparation appears appropriate. Doubling of this dose may provide additional benefit (anecdotal information). The role of a variety of other SSRIs is still being explored. Other compounds such as black cohash and Bellergal®, have been utilized, but these have not undergone placebo-controlled trials to illustrate benefit and toxicities.
Lastly, the use of megestrol acetate for controlling hot flashes can be considered. Megestrol acetate appears to decrease hot flashes as well as does estrogen, at least judged by cross-study comparisons. Many physicians and patients, however, perceive a real risk from using low doses of progesterone in both inducing primary breast cancer in well women and inducing recurrence in breast cancer survivors, since many animal and in vitro models show that progestationals may increase or accelerate breast cancer development and/or progression, and since progesterone is clearly linked with breast cancer etiology in women. As for estrogen, there are no good data to date to demonstrate whether low doses of megestrol acetate used in women with a previous diagnosis of breast cancer increase or decrease the risk of recurrence, or have no effect. Once again, large randomized trials would be required to clarify the safety of this approach. Thus progestationals should probably be regarded with the same degree of caution as estrogen in the setting of a previous diagnosis of breast cancer.
Osteoporosis
Osteoporosis can now be prevented and treated with a number of approaches that do not involve estrogen or progesterone. In addition to recommendations for diet, exercise and calcium supplementation, a wide array of bisphosphonates including didronel, alendronate, clodronate, pamidronate, and residronate are now known to inhibit bone absorption and normalize bone turnover. Alendronate has been studied in large randomized trials and found to improve bone density and reduce fractures [31] in women without breast cancer. Clodronate reduces chemotherapy-induced bone loss in patients with primary breast cancer [32], while residronate also prevents cortical and trabecular bone loss in women with breast cancer who have gone through chemotherapy-induced menopause [33]. In addition, clodronate [34] and pamidronate [35] have significantly reduced skeletal complications and perhaps the development of bone metastases [34, 36] in breast cancer patients. Thus, bisphosphonates clearly provide an alternative for osteoporosis prevention in well women as well as in those with a previous diagnosis of breast cancer.
Tamoxifen also preserves bone density and reduces fractures in postmenopausal women [37, 38], as do a variety of newer selective estrogen receptor modulators (SERMs). Tamoxifen is, however, known to cause a small but significant increase in endometrial cancer and in the risk of deep vein thrombosis (DVT). Raloxifene is one of the newer SERMs which has been recently approved for the treatment and prevention of osteoporosis. It provides somewhat less beneficial effect on bone density than HRT, however, and does not relieve menopausal symptoms [39]. In fact it produces hot flashes, with an incidence similar to that seen with tamoxifen. It does, however, favorably influence total and HDL lipid profiles [40] in a fashion similar to that seen with tamoxifen. Recent follow-up from over 12,000 postmenopausal women randomized to raloxifene versus placebo has also suggested a significantly lower risk of breast cancer in raloxifene users [41, 42]. These data must be interpreted with caution, however, since it was gained from studies in which incidence of breast cancer was not a primary end point. Raloxifene is associated with an increased incidence of DVT similar to that seen with tamoxifen. Preclinical data strongly suggest that raloxifene is not as likely as tamoxifen to cause endometrial cancer, but there are as yet insufficient clinical data to draw a certain conclusion in this regard. A large randomized trial of raloxifene versus tamoxifen as prevention for breast cancer (the National Surgical Breast and Bowel Project STAR trial) is ongoing and will provide considerable additional information on all of these outcomes. It should also be remembered that diet and exercise, and appropriate calcium intake are important factors in the prevention of osteoporosis [43, 44].
Cardiovascular Disease
Similarly CVD can be affected by a variety of other approaches including diet [45], tamoxifen or other SERMS, exercise, hypertension, smoking cessation, and the statins which may reduce total and LDL cholesterol and significantly reduce CVD events [46, 47].
Colon Cancer
Other drugs such as acetylsalicylic acid and nonsteroidal anti-inflammatory drugs, which are without any adverse risk in terms of breast cancer, have also been suggested to prevent colon cancer [48].
Alzheimer's Disease
Drugs or strategies to prevent Alzheimer's and cognitive deterioration are still being sought.
|
LONG-TERM NEGATIVE EFFECTS OF ERT/HRT |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
There has been considerable uncertainty about the role of the addition of progesterone to estrogen in breast cancer risk. Two early studies suggested that the addition of progesterone might reduce breast cancer risk [50, 51], but these studies were small and did not adequately control for confounding. Reliable data on the effects of long-term use of combination therapy have only recently become available [52]. These newer studies provide firm evidence that the addition of progestin to estrogen does not reduce the risk of breast cancer and suggest that the risk is actually increased [49, 53-59]. In the collaborative analysis of epidemiologic studies described above, among current or recent hormone users, the risk of breast cancer was 53% higher for combination therapy and 34% higher for estrogen alone compared with no hormone use [49]. Since the publication of that meta-analysis, at least four subsequent studies have further explained this matter. These important trials are summarized in Table 1
. Once again, all of these studies are observational and subject to the associated risks of bias. Until the results of randomized studies such as WHI are available, however, one must continue to assume that well women receiving ERT/HRT have a small increased risk of developing breast cancer, and that progesterone appears to increase this risk.
|
|
EFFECTS ON MAMMOGRAPHIC SCREENING |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
|
USE OF ERT/HRT IN WOMEN WITH A PREVIOUS DIAGNOSIS OF BREAST CANCER |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
There are many animal and in vitro models in which the development of breast cancer is estrogen-dependent. Virtually every mouse mammary tumor model and mouse xenograft model, as well as many in vitro cell lines, are dependent on estrogen for their growth and spread. Animal and in vitro data concerning progesterone are less conclusive. There are some models in which progesterone has a disease-differentiating effect, while in others it supports breast cancer growth. Some investigators have suggested that estrogen and progesterone may have more of an effect in the development of breast cancer than in its recurrence or metastases [62]. It is well known, however, from the recent Oxford meta-analysis of ovarian ablation in premenopausal women with breast cancer, that ovarian ablation in women with breast cancer results in a significant reduction in recurrence and death [63]. Furthermore, it is felt that the enhanced effects of adjuvant chemotherapy in premenopausal women may in part relate to the induction of ovarian ablation by the cytotoxic drugs involved. There are also, as outlined above, considerable observational data in women suggesting that estrogen and probably progesterone are real contributors to an increased risk of the development of breast cancer.
On the other hand, a large number of lower risk women who are now completing chemotherapy will live for a long time, and are therefore potential candidates for prevention of both long and short-term complications of menopause with ERT/HRT. Patient acceptance of such a strategy is uncertain. In a survey of patient attitudes, Vassilopoulou-Sellin and Zolinski [64] randomly selected 224 women with breast cancer to respond to questions concerning menopause, symptoms related to estrogen deficiency, concerns about osteoporosis or heart disease, and attitudes and perception about ERT. At the time of completion of the survey, 77% were postmenopausal. Of those, 8% had taken ERT at some point subsequent to their cancer diagnosis. Seventy-eight percent were afraid that ERT might precipitate a cancer recurrence, but many were also concerned about the risks of osteoporosis (70%) and heart disease (72%). Forty-four percent of menopausal women indicated that they would consider taking ERT under medical supervision. A survey of a similar population by Couzi et al. [65] reported that 31% would take ERT under medical supervision. These results suggest that the use of ERT/HRT following breast cancer is of interest and concern to women with breast cancer, and that at least some women would consider its use.
There are few clinical data describing the results of ERT/HRT in women with a prior diagnosis of breast cancer. From the observational literature, it has been observed that women who develop breast cancer during pregnancy or have a pregnancy within 1 to 2 years of breast cancer diagnosis, have a poorer outlook than might otherwise be expected [66]. Such studies have also shown that women who become pregnant more than 1 or 2 years following a diagnosis of breast cancer have no obvious increase in recurrence of their disease. It is clear, however, that women who become pregnant following a diagnosis of breast cancer are a highly selected group who may have chosen to become pregnant and/or been advised to consider pregnancy because of a variety of favorable prognostic factors.
At least eight case series of women with breast cancer who have received ERT/HRT for the relief of menopausal symptoms have been published. These are summarized in Table 2.
|
Interestingly, a number of somewhat paradoxical observations have been made regarding the behavior of breast cancer that presents for diagnosis in women currently receiving ERT/HRT. Dhodapkar et al. [71] reported on four women who developed metastatic breast cancer while taking ERT. In each case, withdrawal of ERT alone resulted in regression of metastatic disease. Whether the ERT sped or slowed the development of the metastases was impossible to determine, but the investigators suggested that this maneuver is appropriate as initial treatment for metastatic disease that develops on ERT. Powles and Hickish [72] reported on a single similar case, in which withdrawal of HRT resulted in complete clinical resolution of a primary breast cancer in a patient who deferred primary surgery for 3 months after initial biopsy. However, at surgery, a residual pathologic cancer was present. Booser has also described response to HRT withdrawal as the sole intervention in three of eight similar breast cancer patients [73].
In addition, and also somewhat paradoxically, several observational studies have reported a favorable prognosis for women diagnosed with breast cancer while on HRT [74, 75]. This may relate to the better health care access of HRT users. Recently Melody Cobleigh and others explored the relationship between ERT/HRT and prognostic factors in a cohort of 349 breast cancer patients [76]. A marked increase in the incidence of high S phase was found in women with estrogen receptor (ER)+ tumors who were using HRT at the time of diagnosis, compared to women who had never used HRT and had ER+ tumors. Thus, as HRT may stimulate the growth of receptor-positive cancers, its withdrawal could prove therapeutic, and may explain the improved prognosis reported in some series of such patients.
Goodwin [77] performed a decision analysis of ERT/HRT in women made prematurely menopausal by adjuvant chemotherapy. Based on the available data regarding risk of recurrence, risk of death from other causes, and menopausal symptoms, it appeared that for women with node-negative breast cancer who have substantial menopausal symptoms, the use of ERT/HRT might be reasonable. A decision analysis by Perlman et al. [78], however, suggested that in women with a previous diagnosis of breast cancer, because of the greatly increased relative risk of death from breast cancer in comparison to the risk of death from CVD, osteoporosis or other causes, it would be virtually impossible for such women to gain any overall mortality benefit from the use of ERT/HRT, even if ERT/HRT caused only a very small increase in risk of breast cancer recurrence and death. Thus, it seems that in women with a previous diagnosis of breast cancer, the use of ERT/HRT for short-term symptom relief may be a more appropriate subject for investigation than its use long-term. If short-term use were to prove safe, in well-designed and conducted randomized trials, exploration of the long-term use of ERT/HRT in women at very low risk of recurrence (i.e., ductal carcinoma in situ, very small favorable characteristic invasive disease) might then seem appropriate.
In considering trials of even short-term use, however, it is important to recognize that, just as women with a diagnosis of breast cancer will accept a considerable amount of treatment for very small benefits [79], women with a previous diagnosis of breast cancer are averse to accepting much increased risk of recurrence in order to take HRT [80, 81]. Thus, it is probable that a very large trial would have to be done in order to rule out the very small increases in risk that women would like to avoid. With this in mind, three large studies, the HABITS study (opened in 1996), a second Swedish study (opened in 1998), and a British study (opened in 2000), each randomizing women to ERT/HRT or not for 2 years, after a diagnosis of breast cancer, are now under way. The results of these trials will be greeted with considerable interest. In addition, the smaller randomized trial by Vassilpoulou-Sellin, which will rule out a 10% or greater difference in recurrence rate has been ongoing for more than 8 years [82]. Accrual is not yet complete, however, reflecting the difficulty of carrying out studies in this area.
Until results from these randomized trials are available, it would seem foolhardy to believe that there is no increased risk related to ERT/HRT in this setting. If even the 1.3 to 1.4 relative risk seen in the etiology literature applied to the risk of recurrence, one could see increases in recurrence that would be as high as any gain obtained by giving adjuvant chemo or hormonal therapy. This would clearly be unacceptable.
|
SUMMARY |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMenopausal PhysiologyERT/HRT in Treatment of...Long-Term Positive Effects of...AlternativesLong-Term Negative Effects of...Effects on Mammographic...Use of ERT/HRT in...SummaryReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. W. Goodwin, S. J. Green, C. M. Moinpour, J. D. Bearden III, J. K. Giguere, C. S. Jiang, S. M. Lippman, S. Martino, and K. S. Albain Phase III Randomized Placebo-Controlled Trial of Two Doses of Megestrol Acetate as Treatment for Menopausal Symptoms in Women With Breast Cancer: Southwest Oncology Group Study 9626 J. Clin. Oncol., April 1, 2008; 26(10): 1650 - 1656. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. G. Moorman, A. Berchuck, B. Calingaert, S. Halabi, and J. M. Schildkraut Antidepressant Medication Use for and Risk of Ovarian Cancer Obstet. Gynecol., April 1, 2005; 105(4): 725 - 730. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME |
