Hormone replacement therapy in the menopause: A pro opinion
Dr. Harriet O. Smith MD,
Smith is an Associate Professor and Director of the Division of Gynecologic Oncology in the Department of Obstetrics and Gynecology at the University of New Mexico School of Medicine in Albuquerque, New Mexico
Ours is an aging society. Based on 1991 estimates, 21 percent of the 252 million people living in the United States are older than 55 years; 12.5 percent are aged 65 years or older, 19 million of whom are women. By year 2050, it has been projected that 21.7 percent of our population will be older than 65 years.1 As the average life expectancy of a female born in 1992 is 79 years1 and the average age of the climacteric is 52 years, over a third of the expected life span for women occurs after the menopause,2 when medical diseases become more prevalent.
The recently publicized benefits of hormone replacement therapy (HRT) to reduce the risk of cardiovascular disease and stroke suggest a possible role for HRT for all menopausal women. Recently, even the philosophy of established “contraindications” (a history of breast, endometrial, and/or other estrogen-dependent gynecologic malignancies) has been questioned in view of objective evidence that the protective effects of estrogen may well exceed the theoretical and unproven cancer risks. Clinicians who care for cancer survivors face the difficult challenge of counseling their patients in the absence of prospective, randomized trials that clearly delineate risks. Until guidelines are clearly established, recommendations regarding HRT in cancer survivors, as for all menopausal women, must be individualized.
Rationale for Hormone Replacement Therapy
The principle sex steroid hormones currently administered to menopausal women include estrogens, progestins, androgens, and tamoxifen.
Effects on the Cardiovascular System
Cardiovascular disease (CVD) is the leading cause of death in postmenopausal women in the United States.3 For American women older than 65 years, more than half (55.7 percent) of deaths are attributable to CVD, compared with about one in five deaths from cancer (18.5 percent) and one in four deaths from all other causes.4 A 50-year-old Caucasian woman has a 46 percent lifetime probability of developing CVD and a 31 percent chance of dying from heart disease.3 Prior to the menopause, the risk of CVD is lower in women compared with men within the same age group and is associated with reduced levels of triglycerides, total cholesterol, and low-density lipoprotein cholesterol (LDL-C) levels.5 After the menopause, high-density lipoprotein cholesterol (HDL-C) levels slowly decline, whereas trigylcerides, LDL-C, and total cholesterol levels begin to rise, and the rate of associated CVD increases by tenfold.5,6 Elevations in triglycerides, total cholesterol, and LDL-C and a reduction in HDL-C are all associated with increased CVD risks.7,8
In multiple studies, estrogen replacement therapy (ERT) has consistently been found to favorably affect the lipid profile, with an associated reduction in total cholesterol and LDL-C and increase in HDL-C.7,9–12 The most compelling evidence that unopposed estrogen (conjugated equine estrogen, 0.625 mg/day) significantly elevates HDL-C and lowers fibrinogen levels without adversely affecting blood pressure or serum insulin levels was recently demonstrated in a large, prospective, multicenter, randomized trial.13 Perhaps more importantly, for patients with a uterus, combination estrogen-progestin therapy, especially conjugated equine estrogen with cyclic micronized progesterone (200 mg/day for 12 days per month) was also found to elevate HDL-C without significantly increasing the risks of endometrial hyperplasia associated with unopposed estrogen (one percent versus 34 percent).
Multiple studies have by and large demonstrated a 40 to 60 percent reduction of cardiovascular disease in estrogen users.
However, changes in lipoproteins are not the only postulated mechanism whereby ERT reduces the risk of CVD, as only 25 to 50 percent of the risk reduction for CVD can be accounted for by changes in the lipid profile.11 Estrogen has been shown to increase peripheral blood flow as demonstrated by color-flow doppler and to decrease basal coronary vasomotor tone.7,14–16 Physiologic doses of estradiol-17β potentiates both endothelial-dependent and endothelial-independent vasodilation in menopausal women with impaired vascular function.16 Estrogen may mediate thrombotic factors such as thromboxane, prostacyclin, collagen, and elastin as well as reduce platelet aggregation.3,7
Multiple studies, including case-control, cohort, and cross-sectional, have by and large demonstrated a 40 to 60 percent reduction of CVD in estrogen users.3,11,12,17–25 The relative risks of CVD in estrogen users compared with never users was compared in three independent recent meta-analyses, and CVD risks were consistently reduced over unity (0.55 to 0.65) in estrogen users, regardless of the various endpoints investigated, including fatal and nonfatal myocardial infarction, fatal and nonfatal CVD, coronary stenosis, and sudden death.3,17,18
Effects on Bone Mass
Osteoporosis, characterized by a reduction in bone mass and by an increase in susceptibility to fractures, is an enormous problem. About 25 million Americans are affected, 90 percent of whom are postmenopausal women.26 Hospitalization rates for hip fractures increase exponentially with advancing age, and direct costs exceed three billion dollars each year.27 By 75 years of age, about 50 percent of women will have sustained at least one osteoporotic fracture. Of those with a hip fracture, 34 percent die within six months.26 Risk factors for osteoporosis are outlined in Table 1.
Table Table 1. Risk Factors for Osteoporosis
The lifetime risk of hip fracture equals the combined risks of breast and all other gynecologic cancers.3 It has been estimated that estrogen therapy would reduce the annual mortality rate secondary to hip fractures from 55 to 22 per 100,000 in women in the United States aged 65 to 74 years.28 ERT reduces the rate of bone resorption and thereby retards the development of osteoporosis, reducing the risk of vertebral, hip, and other fractures after six years use by 50 to 60 percent.27,29–32 ERT also reduces the rate of collagen loss from bone and skin. Because collagen is essential for bone strength, reducing the rate of collagen loss reduces the risks of future bone fractures.33
Although less effective than ERT, administration of supplemental elemental calcium (1,700 mg/day) retards bone loss and is a therapeutic option when HRT is contraindicated.34 Other modalities for the prevention of osteoporosis (including exercise, anabolic steroids, norethindrone, calcitonin, fluoride, Vitamin D, alendronate, and others) have variable efficacy and may have unacceptable side effects, increased costs, and/or unknown long-term effects.14,26,35–40 Although reserving ERT for women with decreased bone density as demonstrated by bone densitometry measurements has been proposed,29 serial bone density measurements are an unreliable indicator of fracture risks.40 Gorsky et al,41 using decisional analysis techniques on a theoretical cohort of 10,000 women, found that ERT for 25 years or more would prevent 574 deaths and patients would gain 3,951 quality-adjusted life years (Fig. 1).
Many climacteric women experience symptoms of estrogen deficiency, termed “menopausal syndrome,” including vasomotor instability, sleep disturbances, and psychologic symptoms, such as anxiety, tension, depression, and other psychosomatic complaints.42 The most common complaint relates to vasomotor symptoms, which may last for more than one year in 80 percent of women and up to five years in 25 percent.42,43 Although ERT effectively decreases the frequency and severity of vasomotor symptoms, when estrogen is discontinued, the symptoms recur.44 Progestins, clonidine, alphamethyldopa, anabolic steroids, and belladonna alkaloids have all been used to alleviate menopausal symptoms with less predictable results and often, with unacceptable side effects.45 Several controlled trials examining the affects of estrogen have shown significant improvement in memory, insomnia, anxiety, and irritability, which support a role for estrogen in the relief of some psychologic complaints at the time of menopause.42,46–48 ERT has been found to significantly improve genuine stress incontinence and detrusor instability.49,50 Table 2 summarizes the most common indications for ERT.
The addition of a progestogen for at least 12 days per cycle removes the risk for endometrial adenocarcinoma associated with unopposed ERT.3,51
Effects on the Cardiovascular System
To date, most studies of cardiovascular implications were based upon ERT alone, without progestins.3,17 Because progestins seem to attenuate the benefits of estrogen on the lipid profile, some concern exists that combination therapy will reduce the cardiovascular benefits of ERT.3,23 The potential adverse effects of progestins have been found to be dose dependent and vary with respect to their androgenic properties.7,8 Progestins derived from 17α-hydroxyprogesterone (medroxyprogesterone acetate) and those with little androgenic activity (desogestrel) appear to have less effect on HDL, LDL, and total cholesterol levels than progestational agents derived from 19-nortestosterone.7,8 The 19-nortestosterone derivatives reduce triglyceride and very-low-density lipoprotein (VLDL) cholesterol, whereas the 17α-hydroxyprogesterone derivatives have little or no effect.23 Medroxyprogesterone acetate (Provera) does not reverse the effects of ERT on HDL cholesterol.52 Preliminary data suggests an equal reduction in CVD risk among combined estrogen/norgestrel and unopposed estrogen users.24 Moreover, combination therapy given for 12 months or more was not associated with negative effects upon HDL cholesterol.53–55
Several recent publications evaluating cyclic and/or continuous combination therapy report no adverse lipid profile effects when compared with estrogen alone.12,25,56 In a large, prospective, randomized trial, conjugated equine estrogen (0.625 mg/day) combined with cyclic micronized progesterone (200 mg/day for 12 days per month) had a more favorable effect on HDL-C than conjugated equine estrogen with medroxyprogesterone acetate (2.5 mg daily or 10 mg/day for 12 days per month.)13
Effects on Bone Mass
Combination therapy (estrogen with progesterone) is also effective in the prevention of osteoporosis and may even increase bone mass by promoting new bone formation.37,57–63 Whereas 0.625 mg of conjugated estrogen or equivalent is the lowest effective estrogen dose known to optimally reduce bone resorption,59 combination therapy with low-dose estrogen (0.3 mg) was recently found to decrease the rate of bone loss as effectively as standard estrogen doses (0.625–1.25 mg), suggesting a favorable, synergistic effect on bone mass.60
Other less well known but equally important benefits of progestational agents exist, especially for patients unable to use ERT. The bone-sparing effects of progestin therapy have already been mentioned.37,57,60–63 Oral and parenteral progestational agents have also been used successfully to treat vasomotor symptoms in postmenopausal women, as well as in men undergoing androgen-deprivation therapy for prostatic cancer.52,64–68 Megestrol acetate, in a double-blinded crossover study, was found to reduce the incidence of hot flashes.68 Megestrol acetate also improves appetite and promotes weight gain in patients with anorexia and cachexia associated with chronic illnesses, including advanced cancer.69,70
Although the long-term effects of combination therapy on CVD, osteoporosis, and breast cancer risks require further investigation, even if one assumes an increase in breast cancer risks, combination HRT does not appear to appreciably diminish the protective effects of ERT (Fig. 2).3,28
The safety and therapeutic effects of androgen therapy for menopausal women is uncertain. Open, unblinded and single-blinded studies have demonstrated that androgen replacement enhances libido and results in improved sexual enjoyment.71,72 Adverse effects include acne, voice changes, hirsutism, and clitorimegaly.73,74 Low doses of methyltestosterone do not appear to increase the risk of hepatic damage75,76 or reverse estrogen effects on vaginal epithelium. However, in one study, oral estrogen-androgen combinations were associated with a reduction in HDL, VLDL, and triglyceride levels after six months of use.76 Until sufficient information exists to determine if androgens reduce the protective effects of ERT, androgens should be prescribed with caution.
Tamoxifen is a nonsteroidal, antiestrogenic compound that confers significant protection against breast cancer recurrence and improves survival. Additionally, tamoxifen is associated with a 40 percent reduction in second primary breast cancers of the opposite breast.77 Trials are now underway to evaluate the role of tamoxifen in the prevention of breast cancer development in both premenopausal and postmenopausal patients.78,79 Tamoxifen has estrogen-agonistic properties on bone mineral density, lipid profiles, and the endometrium.80–82 These observations provide support for the concept that tamoxifen alone or with estrogens could be used for HRT.78,80–82
The effect of tamoxifen on the lipid profile fairly consistently demonstrates reduction in total cholesterol and LDL.78,79,81 While most studies do not show a change in HDL levels in tamoxifen users, small, albeit significant decreases in HDL cholesterol have been reported.81 The only available study that evaluated CVD in women using tamoxifen demonstrated a significant decrease in fatal myocardial infarction after five years of tamoxifen use. However, no decrease in other nonfatal cardiovascular events, including stroke or ischemic heart disease, were noted.83
Of greater concern is an increased relative risk for the development of endometrial cancer, similar to that associated with unopposed estrogen use, that has been reported in women taking tamoxifen.84 Also, increased risks for serious embolic events, including pulmonary embolus and deep venous thrombosis requiring hospitalization and/or resulting in death, has been reported in tamoxifen users compared with controls.78 Ocular toxicity (including decreased visual acuity, macular edema, and retinal opacities) has been noted in up to six percent of women after two years use.78,79 Finally, about 15 to 20 percent of women on tamoxifen, regardless of menopausal status, experience vasomotor flushes and/or exacerbations of preexisting menopausal symptoms.79 While tamoxifen probably has beneficial estrogen-like actions on bone and lipid profile, further investigation is needed to determine if the risks of tamoxifen preclude its use as a substitute for ERT.
Combination therapy is effective in the prevention of osteoporosis and may even increase bone mass by promoting new bone formation.
Hormone Replacement Therapy and Risks of Developing Cancer
ENDOMETRIAL CANCER RISKS
At least 35 epidemiologic studies published since 1970 reaffirm the association of unopposed exogenous estrogen therapy and endometrial carcinoma.3,85–91 The risk for endometrial carcinoma is dose dependent and increases with duration of use.3 Although the incidence is increased, mortality is not increased because these tumors are generally of good biologic portent.3,85–90 Also, women in these studies who were on ERT were subject to improved surveillance, which may have resulted in earlier detection of asymptomatic endometrial adenocarcinomas.89 Multiple studies confirm that progestins reduce endometrial cancer risks to baseline or lower (Fig. 2).3,90
Table Table 2. Indications for Estrogen Replacement Therapy
BREAST CANCER RISKS
Several epidemiologic factors support the premise that prolonged exposure to endogenous estrogen is an important risk factor in breast cancer development. For example, early menarche, late menopause, low parity,92,93 and obesity (increased central to peripheral body fat distribution)94 are all associated with increased breast cancer risks, whereas oophorectomy,93 vigorous exercise (which presumably alters ovarian function),95 early childbearing,93 and breast feeding for a minimum of three months have the opposite effect.96
Because pregnancy is characterized by dramatic elevations in endogenous steroid hormones including estradiol, estrone, progesterone, human chorionic gonadotropin, and prolactin, if exogenous estrogen administration increases the risk for breast cancer development, one might argue that pregnancy (particularly at a young age) and multiparity should increase breast cancer risks. In reality, the association between pregnancy and breast cancer is far more complex. The protective effect of a first pregnancy before 20 years of age has clearly been established, and most reports have found that breast cancer risks are further reduced with each subsequent pregnancy. Nevertheless, the protective effect of parity is lost in elderly primigravidas (>35 years) compared with nulliparous women. Moreover, uniparity in women older than 35 years compared with nulliparity has been found to increase breast cancer risks (odds ratio, 1.26; 95 percent confidence interval, 1.10–1.44).97
Pregnancy is associated with a transient increase in the short-term (15 years or less) risks of breast cancer, followed by a reduction in long-term breast cancer risks compared with age-matched nulliparous women. This supports the hypothesis that pregnancy has opposing influences on breast cancer risks. Pregnancy stimulates the differentiation of stem cells, which renders them resistant to mutations and long-term protection is conferred. However, pregnancy increases the short-term risks of breast cancer by inducing the growth and malignant transformation of cells in the early stages of carcinogenesis.97 This latter effect may explain why elderly primigravidas are at increased risk for breast cancer development.
In the laboratory, estrogen promotes the growth of some breast cancer cell lines, and in high doses can induce mammary tumors.98,99 Because of these effects, as well as the apparent association between breast cancer development and uninterrupted or prolonged estrogen exposure, it has been assumed that estrogen induces breast cancer development in women. However, certain biologic factors suggest that estrogen-associated tumors are of better biologic portent. For example, positive status for estrogen receptor and progesterone receptor is predictive of improved outcome.
ER and PR status continues to be used to predict which patients are likely to respond to hormonal manipulation, although the correlation between receptor status and response to such therapy is imprecise (about 60 percent of breast cancers have significant levels of ER; however, only two thirds of ER-positive tumors respond to hormonal manipulation and five to 10 percent of ER-negative tumors respond to endocrine therapy).98 The mechanism whereby estrogen stimulates tumor growth is poorly understood. However, estrogen is involved in the regulation of several genes necessary for DNA replication and stimulates the expression of a number of growth factors such as insulin-like growth factors and transforming growth factor-α, as well as the nuclear proto-oncogenes c-fos, c-myc, c-myb, and c-jun.98,99 High levels of expression of the transmembrane protein c-erbB-2 and overexpression of the receptor for epidermal growth factor are associated with a poor outcome, and often, with negative ER status, suggesting that these oncogenes mediate cancer cell growth by autocrine and paracrine mechanisms that have become independent of estrogen.
Despite the fact that about 50 epidemiologic studies and reviews of studies have been published since 1971, there is still no consensus regarding postmenopausal estrogen use and the risk for breast cancer.100–118 Most of these reports are case-control studies, and most have found no association between breast cancer risk and ever-use of estrogen compared with never-use. However, increased duration of use does appear to increase breast cancer risks.101,106–108,114 Interpretation of these studies is difficult for a number of reasons, including the following: (1) in many case-control studies, hospitalized patients were used as controls; (2) estrogen prescription practices have changed over time from higher doses of conjugated estrogen to lower doses and combination therapy; (3) patients with surgical menopause (whose risk for breast cancer is already reduced) are not evaluated separately from patients undergoing natural menopause; (4) premenopausal and postmenopausal breast cancer victims are not analyzed separately; (5) most studies lack long-term follow-up; (6) the dose of estrogen is unknown; and (7) many confounding variables (parity, obesity, age of first pregnancy, and oral contraceptive use in patients and/or controls) are not assessed.
Despite publication of numerous epidemiologic studies, there is still no consensus regarding postmenopausal estrogen use and the risk for breast cancer.
In an attempt to provide dose-response estimates of the risks for breast cancer, four meta-analyses of HRT and breast cancer have been performed and are summarized in Table 3. For unstated reasons, two of these authors included only case-control studies. Using a questionable exponential curve-fitting process to extrapolate long-term risks for estrogen, two of these authors found significantly higher risks for breast cancer with extended use (12 or more years).101,104 Two of these studies found no increased risks with combination therapy or low-dose ERT.103,104 Furthermore, the use of conjugated estrogens in doses of 0.625 mg/day or less for up to five years use was not associated with an increased risk of breast cancer development.101,103
Seven prospective studies have been published assessing the relation between menopausal estrogen use and breast cancer risk (Table 4).108–114 Colditz et al114 extended their follow-up of the participants in the Nurses' Health Study published in 1992 to include 1,935 cases of newly diagnosed invasive breast cancer during 725,550 person-years of follow-up. A significant increase in the risk of breast cancer was found among women taking estrogen or combination HRT as compared to postmenopausal women who had never used hormones; the relative risk was greatest in women aged 65 to 69 years. However, there was no increased risk for breast cancer among women who had used hormonal therapy for less than five years.
It has been suggested that ovarian stimulation with its associated increases in endogenous estrogen and progesterone may be a risk factor for breast cancer development, lending additional support to the premise that ERT may increase breast cancer risks.115 However, when the preliminary report of 16 cases of breast cancer in patients undergoing ovulation induction is carefully examined, it becomes apparent that other proven risk factors such as advanced age at first birth and nulliparity may fully explain the increased risks. Over half of cases were nulliparous or first gave birth at 30 years of age or older.
Table Table 3. Summary of Metaanalysis of Effects of Estrogen Replacement Therapy on Breast Cancer Risks
Even if one assumes a slight increase in breast cancer risks, ERT, by reducing the risks of cardiovascular disease and by preventing or attenuating osteoporosis, significantly improves life expectancy. ERT for at least 15 years has been found to reduce combined mortality risks by 40 percent.51
In the endometrium, progestins reduce epithelial cell growth mediated by estrogens and stimulate differentiation of estrogen-primed cells, thereby reducing the risks of endometrial hyperplasia and carcinoma.3,119 Additionally, the progesterone-dependent enzyme 17β-hydroxysteroid mediates the conversion of estradiol to its less active metabolite, estrone.116
In the breast, however, there are concerns that exogenous progestins may increase the risks of breast cancer development or recurrence.117,118,120,121 This is based upon the fact that progesterone-induced DNA synthesis and mitotic activity in the breast reaches its peak during the progesterone-dominant luteal phase of the menstrual cycle.122 Nevertheless, there is no consistent evidence to support the premise that an association exists between progesterone and breast cancer development.117,118,120,121,123
At least one large epidemiologic study suggests that progestational agents significantly reduce the risk of breast cancer.123 Although more recent epidemiologic studies have not demonstrated a protective effect from progesterone, combination estrogen-progestin therapy has not been found to be associated with an increased incidence in breast cancer, especially when limited in use to less than five years.113,114 This is supported by studies of the more potent progestins used as contraceptives. The World Health Organization collaborative study of the progestational contraceptive depot-medroxy-progesterone acetate (DMPA), found that DMPA with extended use was not associated with increased breast cancer risks, except possibly in users younger than 35 years.124 Studies of progestin effects on breast cancer cell lines have demonstrated both proliferation and inhibited cell growth.122,125,126 Progestins are cytotoxic to estrogen-primed MCF-7 cells, a human adenocarcinoma cell line that contains both ER and PR.126 Although studies of hormone effects on benign breast tissue in culture are limited, progestins in one study were found to inhibit epithelial cell multiplication.127 Finally, pharmacologic doses of progestins have been used successfully to treat postmenopausal breast cancer patients and are considered by some the agent of choice after tamoxifen failure.128
ERT for at least 15 years has been found to reduce combined mortality risks by 40 percent.
Risks of HRT in Cancer Survivors
HRT FOLLOWING GYNECOLOGIC CANCER
Despite concerns that hormonal therapy may induce tumor growth and metastasis, there are no clinical data to support the rationale of withholding ERT in women with a previous gynecologic malignancy, especially carcinoma of the cervix, vulva, gestational trophoblastic disease, or ovary, with the possible exception of the endometrioid epithelial ovarian cancer.129 Data on the effect of ERT on ovarian cancer are scarce; the only available study reports no increased risks.130
Table Table 4. Prospective Studies Evaluating The Effects of Hormone Replacement Therapy and Breast Cancer Risks
In contrast to other gynecologic tumors, ERT for patients with a history of endometrial adenocarcinoma remains controversial, and literature necessary to make informative recommendations is scarce. In 1985, Creasman et al131 presented the first evidence that ERT did not increase local or systemic recurrences in low-risk patients (FIGO clinical stage 1). This study was a retrospective comparison of patients who did or did not receive ERT, and both vaginal and oral estrogens were used. However, established risk factors for recurrence, including grade, depth of invasion, positive nodal disease, positive cytology, and positive hormone receptor status (ER or PR) were similar in both groups. After controlling for these known risk factors, patients receiving ERT were found to have a significantly longer disease-free interval.131
These results are corroborated by Lee et al132 who treated 44 selected patients (low-grade lesions, less than 50 percent local invasion, and without nodal metastasis) with oral estrogen replacement. In a follow-up interval of 64 months, there were no recurrences or intercurrent deaths among the ERT recipients. Interestingly, five of the eight patients dying of intercurrent disease in the 99 patients not receiving ERT died from myocardial infarction, suggesting a survival advantage for cancer patients treated with ERT. Finally, in a preliminary report, Baker et al133 have followed 31 low-risk endometrial cancer survivors who were treated with oral estrogen, and there have been no recurrences to date.133
Pending additional information from ongoing studies, Buller129 has proposed the following guidelines for HRT in endometrial cancer survivors. ER and PR status should be determined at the time of surgical staging. Patients at low risk of recurrence may promptly begin ERT. Also, patients at high risk for recurrence who are ER negative should be offered ERT. Patients who are ER positive and are at high risk of recurrence should probably be observed for a two- to five-year disease-free interval prior to initiating therapy, and consideration should also be given for adding a progesterone to those patients who are PR positive.129
BREAST CANCER SURVIVORS
It is estimated that 184,300 women in the United states will be newly diagnosed for breast cancer annually, and 44,300 women will die from their disease.134 A dramatic rise in the rates of breast cancer, mostly of noninvasive and early-stage tumors, occurred in the United states between 1982 and 1987, presumably secondary to improved screening.117 The five-year survival rate for women with localized breast cancer has risen from 78 percent in the 1970s to 93 percent today.134 More than 500,000 menopausal women will survive for at least five years after diagnosis of breast cancer, and 20,000 women annually reach menopause with an excellent prognosis after breast cancer therapy.135,136
Postmenopausal women with breast cancer and premenopausal patients developing ovarian failure subsequent to adjuvant chemotherapy are almost universally denied ERT based on the belief that estrogens are responsible for the development of breast cancer and that ERT induces growth in occult metastatic disease.102,117,118,135,136 There are a number of observations, however, that have been used to support the argument that physiologic doses of ERT are not harmful for breast cancer survivors.118
ERT for patients with a history of endometrial adenocarcinoma remains controversial.
Table Table 5. Estrogen Replacement Therapy Regimens Currently Used in Clinical Practice
First, in a number of studies, women diagnosed with breast cancer during or shortly after pregnancy have the same prognosis as matched, nonpregnant breast cancer patients.137–139 Most studies that have shown no appreciable effect of pregnancy on breast cancer survival have combined data from concurrent or recent previous pregnancies and subsequent pregnancies to obtain adequate numbers of patients for analysis. In young women (ages 20 through 29), concurrent or recent pregnancy has also been shown to be associated with an adverse outcome.140 All of these reports are limited in their ability to assess the effect of pregnancy by sample size and by the fact that young women generally have biologically more aggressive tumors and a poorer outcome.141,142
Second, pregnancy after successful treatment for breast cancer does not adversely affect survival. Several studies have found improved survivals,118 while others have found reduced survivals in these patients.143
Third, there is no association between oral contraceptive use and increased breast cancer risks.100,133 Finally, a link between HRT and breast cancer has not been established.101–118 In fact, the prognosis of breast cancer for patients receiving HRT prior to diagnosis has been consistently superior to that of women with no previous exposure.106,123,144,145
Bergkvist et al,144 in a case-control study of 261 breast cancer patients who had used estrogen or estrogen-progesterone combination therapy prior to their diagnosis of breast cancer, found a 40 percent reduction in excess mortality among these patients compared with 6,617 breast cancer patients without prior estrogen therapy. The favorable outcomes were found only in women older than 50 years at diagnosis. Bonnier et al145 found that women who developed breast cancer while on HRT had fewer locally advanced cancers, more well-differentiated tumors, and an improved metastasis-free survival.
Leading experts in extensive reviews of the potential risks of hormonal therapy weighed against the proven benefits, have strongly argued for a reevaluation of the dictum that breast cancer survivors should be denied ERT.102,105,117,118,146,147 A prospective study of HRT, including ERT with progesterone or tamoxifen, in select breast cancer survivors is long overdue. In a recent survey of the attitudes of women with breast cancer, patients were equally concerned about the protective effects of HRT and the risks for cancer recurrence; 71 percent of patients with early-stage disease expressed an interest in taking ERT under careful medical supervision.146 For the patient at risk, the American College of obstetrics and Gynecology advocates prescribing ERT on an individual basis after fully informing patients of the known benefits and possible risks.147
The prognosis of breast cancer for patients receiving HRT prior to diagnosis has been consistently superior to that of women with no previous exposure.
HRT FOLLOWING NONGYNECOLOGIC CANCERS
Currently there is no evidence to suggest that HRT increases the risks for recurrence of other nongynecologic solid tumors, including gastrointestinal cancers, lymphomas, or leukemias.129 The role of ERT in patients with a history of melanomas is uncertain.145
Hormone Replacement Therapy: Indications, Regimens, and Surveillance
Table 5 summarizes recognized indications for ERT. The benefits of ERT in reducing the incidence and mortality from cardiovascular disease may exceed all of the other protective effects combined and support the practice of prescribing ERT to otherwise asymptomatic postmenopausal women.19,20,37,41,59
HRT schedules currently advocated are summarized in Table 5, along with the recognized limitations and benefits of proposed schedules of administration. Continuous ERT, oral and transdermal, is becoming increasingly popular because of ease of administration, reduced withdrawal bleeding, and improved compliance.148,149 Unopposed ERT is recommended for patients who have had a hysterectomy because the long-term effects of progestins on the breast are yet to be defined.2 For women with a uterus, regardless of route of administration, the American College of obstetrics and Gynecology recommends the addition of progesterone for 10 to 12 days per month to reduce the risk for endometrial cancer.2 Combined estrogen/progestin therapy can also be prescribed. Because of potential adverse effects upon lipoprotein levels, C-21 progestins are generally advocated over 19-norprogestins.7,8
Prior to initiation of therapy, careful history, physical examination, and laboratory evaluation (including cervical cytology, measurement of blood pressure and lipid profiles, and screening mammography) should be performed.2 it has been recommended that a progesterone challenge test be considered for all postmenopausal women with an intact uterus under evaluation for ERT. Withdrawal bleeding identifies the presence of an estrogen-primed endometrium and signifies the need for an endometrial biopsy to rule out an occult endometrial cancer.37 Alternatively, this procedure may be reserved for patients at greatest risk (e.g., morbidly obese, anovulatory cycles, etc).
Compliance must be stressed. In a recent study, 40 percent of patients with known low bone mineral density were not using therapy by their eighth month follow-up.149 Women generally discontinue therapy because of nuisance side effects such as breast tenderness or withdrawal bleeding and, often, because of fear of cancer.150,151 As one might anticipate, compliance is significantly enhanced when the individual is provided with a thorough explanation of the purpose of hormone replacement, its side effects, and potential risks.151,152
The preponderance of data support the benefits of HRT in estrogen-deprived and menopausal women to reduce the risks of osteoporosis and cardiovascular disease and to enhance quality of life and life expectancy. Controversies exist with regards to the risk-benefit ratio in women with a history of estrogen-dependent gynecologic tumors or breast cancer. Until these issues are resolved, physicians must carefully weigh, on an individual basis for each patient, the potential risks against the known benefits. Women should be counseled regarding the benefits of exercise, weight control, breast feeding, and cessation of cigarette smoking or excessive alcohol to reduce their risks of cancer, cardiac disease, and/or osteoporosis.92–96 HRT is not a panacea for an unhealthy lifestyle. When ERT is contraindicated, viable alternatives to retard bone loss and/or control vasomotor symptoms include calcium supplementation and progestin therapy. The role of tamoxifen as an alternative HRT in women at increased risk for breast cancer development is currently under investigation.