Proceedings of a ConferenceHeld at the Boar's Head Inn, Charlottesville,Virginia, September 21-23, 1997
Problem: Several million women worldwide have survived breast cancer but are currently advised against the use of estrogen for the management of menopausal symptoms and for the prevention of early cardiovascular death and osteoporosis.
Consensus Conference: A unique meeting involving international experts and breast cancer survivors was convened to address this problem.
Use “tailored treatment strategies,” which avoid the use of estrogens while providing its benefits (short and long term), to address individual patients’ needs.
Perform research trials to evaluate estrogens or estrogen alternatives in selected groups of women in whom the benefits might potentially outweigh the risks.
Conduct clinical trials to exploit the highly favorable properties of selective estrogen receptor modulators (SERMs).
Forge a “Partnership for Progress” between patient advocate groups and health professionals in order to facilitate research and education about treatment options.
Further develop the “Partnership for Progress” by exploring the establishment of a patient-initiated registry to determine what alternatives or standard medical approaches patients are using to manage estrogen deficiency symptoms and to prevent cardiovascular disease and osteoporosis.
Definition of the Problem
Increased patient awareness, mammography screening, and the use of adjuvant therapy have resulted in earlier diagnosis of breast cancer and a greater probability of long-term survival. Consequently, a large and increasing number of women who have survived breast cancer are alive. In two-thirds of these patients, the onset of menopause occurred prior to the diagnosis of breast cancer. In many of the others, ovarian failure resulted from adjuvant chemotherapy or occurred spontaneously. A large fraction of these women currently experience symptoms of estrogen deficiency and/or can expect to develop premature heart disease and osteoporosis.
At present, estrogen replacement therapy is considered by many to be contraindicated in these menopausal breast cancer survivors since estrogens may accelerate the growth of occult metastases. How to treat the range of problems related to estrogen deficiency in these patients is largely unexplored at present. Both the short-term effects of estrogen deficiency, such as vasomotor instability and urogenital atrophy, and the long-term consequences, such as osteoporosis and heart disease, represent important health and quality-of-life issues for these breast cancer survivors.
There are several million breast cancer survivors worldwide. Specifically, in the United States, 180,000 women were diagnosed with breast cancer in 1997. Approximately 97,000 of these women have an extremely low chance of experiencing a recurrence of their cancer during their lifetime. With an average age at diagnosis of 60 years and a 25-year expected survival, the current number of breast cancer survivors in the United States may approach 2.5 million women. Since breast cancer is now being detected at an earlier stage than previously and since adjuvant chemotherapy may cause ovarian failure, an increasing number of women are postmenopausal at a younger age after breast cancer treatment.
This conference was convened to consider how menopausal breast cancer survivors should be treated at the present time and what future studies are needed to develop improved therapeutic strategies.
Patient advocates, as well as experts from a wide range of disciplines, including medical oncology, surgery, gynecology, endocrinology, radiology, nursing, epidemiology, and the basic sciences, were represented (see list of conference participants). The conference planners wished to fully integrate women with a previous diagnosis of breast cancer into the schedule of formal talks and discussions so that the perspective of the patient would receive appropriate emphasis.
The conference focused on three specific areas and attempted to reach consensus or identify areas of divergent opinion in each. The first addressed the question of initiating clinical trials with estrogen replacement therapy in subsets of women surviving breast cancer or using estrogens prior to the completion of trials. The second topic evaluated the potential for the use of selective estrogen receptor modulators (SERMs) in treating the problems arising from estrogen deficiency. The third considered the use of surrogates for estrogen to treat specific problems related to estrogen deficiency.
Key aspects of each of these topics are considered in this report, and areas of consensus and divergence are described under each topic. In addition, a consensus statement was prepared by the patient advocates to reflect their unique perspective on the various issues discussed.
Menopausal women with a previous diagnosis of breast cancer, like other women, experience a variety of hormonal changes that potentially affect every aspect of their lives. In addition, many of the therapies currently recommended following a diagnosis of breast cancer produce body changes that can worsen this situation. Lumpectomy, breast irradiation, mastectomy, and axillary dissection each produce changes in body and body image. Most combination chemotherapies produce either complete menopause or at least some degree of ovarian dysfunction. Menopause can be produced abruptly in this situation, precipitating acute menopausal symptoms, which add to the anxieties, symptoms, and concerns already associated with the diagnosis of breast cancer and its surgical, chemotherapeutic, or hormonal therapy.
A traditional belief of the medical profession holds that estrogen and/or progesterone therapy represents an unacceptable risk in women surviving breast cancer. This belief is not unreasonable, as it is based on much of our knowledge about the causes and treatment of breast cancer. Estrogen and progesterone exposure are closely related to the development of breast cancer. In established breast cancer, removal or reduction of estrogen often results in shrinkage of breast cancer or in prevention of recurrence. Thus, both physicians and patients remain extremely cautious about the routine clinical use of estrogen or progesterone in women who have ever had a diagnosis of breast cancer.
On the other hand, recent studies in patients without breast cancer suggest that estrogen replacement therapy can lengthen life. Thus, it is possible that withholding estrogen from women with a previous diagnosis of breast cancer could increase their mortality from cardiovascular disease. Small observational studies in women with breast cancer receiving estrogen replacement therapy have not shown more rapid recurrence, but properly randomized studies have not yet been conducted. Thus, important information is lacking regarding the safety and benefits of estrogens in women surviving breast cancer.
Consideration of Clinical Trials
The conference participants considered whether any trials of estrogen replacement should be undertaken in survivors of breast cancer, and, if so, in which subset of patients. The participants agreed that the ability to control menopausal symptoms with surrogates for estrogen, while effective in some patients, was limited in others.
The majority of polled participants, but not all, agreed on the need to conduct trials of estrogen replacement therapy in selected groups of women surviving breast cancer. There was agreement that the currently ongoing trials, such as the Hormone Replacement Study After Breast Cancer: Is it Safe? (HABITS) trial, a large multi-institution Scandinavian and European trial coordinated from the Uppsala University by Dr. Lars Holmberg, and other trials will not provide all of the information needed.
Those favoring clinical trials of estrogen replacement believe that an answer to this question is required and that it would not be ethical to continue to make recommendations to patients without greater scientific evidence. Those arguing against clinical trials believe that major difficulty will be encountered in entering a sufficient number of patients in these trials to answer the safety questions with acceptable statistical power. One participant suggested that information regarding safety might be gained from case-control observational studies resulting from the establishment of a patient registry.
Most agreed that the initial trials should consist of short-term studies focusing on relief of symptoms of vasomotor instability and urogenital atrophy but not on prevention of osteoporosis or heart disease. Safety issues under these circumstances would involve the risk of accelerating the growth of occult metastases but not the initiation of new second primary breast cancers.
One group favored trials in patients at lowest risk of adverse effects from estrogens, namely, women with estrogen receptor (ER) negative tumors. Other participants favored trials in women with receptor positive tumors, arguing that studies in these women would provide stronger evidence of safety if recurrences were not increased. The latter women were also thought to provide higher statistical power to detect significant differences in recurrence. The pros and cons of these two approaches were felt to represent a dilemma with no easy resolution.
Review of an ongoing trial and strong opinions expressed by patient advocates suggested that only a small fraction of breast cancer survivors would accept the use of estrogen replacement therapy, even if studies suggested relative safety. Consequently, trials with other approaches designed to relieve menopausal symptoms should also be carried out in order to develop safe, acceptable alternatives for women. Some participants felt that the use of progestins might not be safe in this setting, even though megestrol acetate is known to be an effective treatment for advanced breast cancer, albeit at higher doses than those used for vasomotor instability.
The participants discussed at length, but could not agree on, specific groups of women to be involved in initial trials of hormone replacement therapy (HRT). Many reasoned that women undergoing chemotherapy-induced menopause experience particularly severe symptoms and should be targeted for initial trials of hormone replacement. Most agreed that such trials should commence only after subsidence of chemotherapy-related symptoms in order to avoid confounding the interpretation of results. Patient advocates and others expressed the opinion that women in this category would be most frightened of the adverse effects of estrogen and that accrual into such trials would be too small to obtain meaningful information.
Substantial discussion addressed clinical trials of the combined use of tamoxifen (Nolvadex) and replacement estrogen in patients with ER positive tumors. At the conference, concepts had been formally presented regarding the stoichiometry between tamoxifen and estrogen for the ER and the differential agonistic and antagonistic effects of tamoxifen on various target tissues.
It was noted that tamoxifen is an effective antitumor agent for advanced breast cancer in cycling premenopausal women with estradiol levels of 1,000 to 2,000 pmol/L. Based on this observation, tamoxifen should remain an effective antitumor agent in postmenopausal women given small amounts of replacement estradiol sufficient to increase plasma levels only to the 150- to 450-pmol/L range. Under these conditions, the effects of estrogen might relieve hot flashes and symptoms of vasomotor instability without stimulating tumor growth. Preliminary biochemical data were presented to the participants regarding patients receiving therapy with both tamoxifen and conjugated estrogens (Premarin).
Based on this information, the participants believed that the combination of tamoxifen with estrogen or progesterone might potentially relieve menopausal symptoms without increasing the risk of tumor recurrence. It was agreed, however, that the data presented were insufficient to conclude that symptoms would be fully relieved by this approach.
Most participants agreed that small pilot studies to determine the efficacy of this approach in relieving symptoms should be followed by large, randomized, controlled trials to ensure safety. This approach was favored particularly for women with ER positive tumors.
The conference participants initially attempted to design prototype clinical trials of HRT during the consensus- building period. This was found to be impossible and, as expressed by several discussants, not the purpose of the consensus conference.
The panel then agreed to establish general principles on which such trials could be based. A consensus was reached that trials of hormone replacement should initially involve women who are symptomatic, rather than women in whom the prevention of osteoporosis or heart disease is the primary goal. Trials should be short term to minimize concerns about stimulation of occult micrometastases. Only with long-term estrogen replacement would the initiation of new second primaries be an important consideration.
Groups of women selected for such trials should have findings suggesting that the benefits of HRT are likely to outweigh the risks. These might include women receiving tamoxifen; patients with small, node-negative or low-histologic tumor grades, in whom the likelihood of long-term survival is great; women with receptor negative tumors; and women with a long disease-free survival before treatment with estrogen.
Use of Estrogen Replacement Prior to Completion of Clinical Trials
The participants considered at length whether it might be appropriate to offer HRT to selected women, as an interim measure, prior to the completion of clinical trials. All agreed that other established means of controlling symptoms or preventing osteoporosis or heart disease should be utilized before considering estrogen therapy.
In those women who do not respond, entry into a clinical trial would be the preferable approach. However, nearly all agreed that a subset of women continue to experience severe problems from estrogen deficiency that might only be controlled by HRT. The participants agreed that an informed woman, knowing all the potential benefits and risks of estrogen, could choose to take estrogen and may be supported in that decision. Under those circumstances, informed consent by the patient should precede the use of estrogens (most, but not all, felt this should be written informed consent). The unknown but potential risks of stimulating occult metastases or of causing a new cancer should be fully discussed.
A clear distinction was made that there was not a consensus to recommend estrogen in selected patients, but rather, that a woman is free to make her own decision provided she is fully informed. Estrogens or progestins in this setting should be prescribed in the lowest doses, for the shortest duration of time, and only after full discussion. This strategy considers the informed patient as the final decision-maker. The heath care provider serves to guide the patient through the difficult process of assessing known and unknown risks and benefits.
To capture the essence of the discussion, a comprehensive statement was offered to the participants and agreed on:
“In women who have had an established diagnosis of breast cancer, we should seek other established symptomatic or health promoting interventions before considering the use of estrogens. When estrogen is used as a last resort, it should be used in the lowest dose for the shortest duration of time and only after full discussion of concerns regarding potential risks with respect to breast cancer outcomes. When estrogen is being considered, the role of the informed woman as the final decision maker should be accepted by the health care practitioner.”
Summary of Consensus Points
Studies of a variety of methods to control the short-term effects of estrogen deficiency in breast cancer survivors are necessary. One approach is to examine the efficacy of surrogates for estrogen. The other is the use of estrogen itself, where the primary goal should be to examine the efficacy and safety of estrogen and progesterone replacement therapy.
Initial emphasis should be on trials for relief of menopausal symptoms, including hot flashes, vaginal dryness, urinary symptoms, and painful intercourse.
Carefully designed trials to explore the effectiveness and safety of estrogens and/or progesterone in women with a previous diagnosis of breast cancer who are also receiving tamoxifen should be undertaken.
Well-designed trials, starting with smaller pilot studies to determine efficacy and followed by large, randomized, controlled trials to ensure safety, will be required to establish the indication for combined estrogen and antiestrogen therapy.
Since only a small fraction of women surviving breast cancer will accept HRT, trials of alternative therapies to relieve symptoms are required.
A series of principles intended to guide initial clinical trial design included the following:
Clinical trials should include only symptomatic women and HRT given over the short term.
Well designed, randomized trials are required.
Studies should involve women in whom the benefits of HRT are likely to outweigh the risks. Specifically, this might include: women with small, node-negative or low-histologic grade tumors; women receiving tamoxifen; women with receptor negative tumors; and women with a long disease-free survival.