Tamoxifen has well-documented activity in reducing breast cancer mortality. In addition, it has several secondary
ABSTRACT: Tamoxifen has well-documented activity inreducing breast cancer mortality. In addition, it has several secondaryeffects that may have an overall impact on long-term survival, and thatare of particular importance in patients surviving the primary breast cancer.To fully understand the therapeutic role of tamoxifen, it is importantto take into account survival differences related to all conditions knownto be affected by tamoxifen. The purpose of this report is twofold. First,it provides information on mortality rates for conditions other than primarybreast cancer in patients who survive the primary breast cancer, and theeffect of tamoxifen on these mortality rates. Second, it outlines the principalside effects of tamoxifen that can affect quality of life and briefly summarizestheir management. [ONCOLOGY 11(Suppl 1):45-48, 1997]
The main objective of this article is to describe the effects of adjuvanttamoxifen (Nolvadex) on overall (net) mortality in patients surviving primarybreast cancer, and in particular, to review the interactive relationshipbetween the effects of tamoxifen on mortality and its effects on qualityof life. Particular attention is paid to the issue of quality of life andits management, emphasizing that improvement of quality of life resultsin increased compliance, which, in turn, may improve the overall survivalimpact of tamoxifen. Quality of life represents an important, but not yet sufficiently explored consequence of tamoxifen therapy. Hence, a thoroughand unbiased quantitative analysis of mortality outcomes is essential,as mortality reduction may drive parallel efforts to improve quality oflife in order to ensure compliance.
The estrogen agonistic and antagonistic effects of tamoxifen are welldocumented.[2] These actions of tamoxifen produce a range of benefits andside effects that may alter mortality from other conditions affecting breastcancer patients treated with tamoxifen. The changes in mortality may occurin opposing directions and vary in order of magnitude.
Current evidence from several randomized trials demonstrates that tamoxifenreduces the incidence of contra- lateral breast cancer and decreases cardiovasculardisease incidence and mortality with relative risk (RR) reductions of 0.5-0.8.[3-6]In contrast, uterine cancer and possibly thromboembolic episodes may resultin excess deaths, with relative risk increases of 2-7.5 (Table1).[4,7-9]
A recent British Columbia report[1] expressed the long-term impact oftamoxifen on the overall (net) mortality in breast cancer patients survivingtheir primary disease in quantitative terms taking into account all knowntamoxifen- associated conditions that have the potential to affect latemortality. An attempt was made to integrate age- matched mortality ratesof tamoxifen-associated conditions as they interact, in order to avoida potential bias that could result if only one of the conditions was considered.
Our calculations estimated that despite a projected modest increasein mortality from uterine cancer and thrombo- embolic episodes, there willbe a more substantial mortality reduction due to avoidable deaths fromcontralateral breast cancer and cardiovascular events. These calculationswere in addition to tamoxifen's previously confirmed improvement of breastcancer-specific survival, due to avoidance of locoregional and systemicrelapses from primary breast cancer.
Although recent reports of increases in the incidence of uterine cancerassociated with tamoxifen usage [7,8] have captured the interest of thescientific community and media, our calculations show that there is onlya moderate excess of deaths from endometrial cancer and that there is amore substantial reduction in deaths from contralateral breast cancer andcardiac events. One of the main conclusions of our integrated age-matchedanalysis is that the rather modest relative risk reductions for conditionssuch as contralateral breast cancer or cardiac events will result in largeabsolute numbers of saved lives, because the life-long underlying mortalityrates for these conditions are high. On the other hand, using even veryhigh relative risk increases for conditions such as uterine cancer or thromboembolicepisodes results in a lower absolute number of excess deaths, because theunderlying population mortality rates for these conditions are low.
Because the mortality effects of different conditions are exerted inopposite directions, the overall picture of tamoxifen's impact on long-termsurvival could be skewed if analyses are restricted to only one conditionat a time. Thus, our methodology and results emphasize that only an integrated,age-matched analysis that captures all conditions simultaneously will havethe potential to present an overall mortality picture and take into considerationall treatment-affected conditions. With this type of approach, the frequentlyobserved unintentional or intentional reporting bias could be reduced oravoided.
While fear of uterine cancer is the most frequently cited reason forthe reluctance of physicians and patients to consider tamoxifen therapy,it is the less well quantitated area of quality of life that constitutesthe main problem with tamoxifen compliance. Tamoxifen-associated side effectsrange from hot flashes, vaginal dryness, and associated dyspareunia withreduced libido--the more typically described antiestrogen effects of tamoxifen--tomuch less frequently discussed, yet commonly observed complaints, suchas insomnia, depression, ocular irritations, gastrointestinal symptomatologywith nausea, and weight gain.
Although reported frequently in mild forms, the frequency of side effectsof tamoxifen affecting quality of life is less well documented becausequantification of quality of life is hampered by the subjective aspectof the complaints. The evaluation of quality of life is made more difficultbecause most clinicians do not use validated quality-of-life instrumentsconsistently. Such instuments could provide reasonably objective evidencefor the magnitude of effect on quality of life,[21,22] particularly ifthey were used more frequently and prospectively in tamoxifen studies,with testing of quality of life as a defined objective.
Despite tamoxifen's definitively proven survival benefit, its adverseeffects are sufficiently significant that they are an issue that needsto be addressed. Of particular importance is evidence that side effectsaffecting quality of life, which could be tolerated in patients with advancedor developed breast cancer, represent a large-scale problem in women whoare otherwise well and who may be considering taking tamoxifen for breastcancer prevention.
From this perspective, it cannot be overemphasized that while the ultimategoal of therapy is overall improvement in survival, quality-of-life issuesare of great importance, not only because of their impact on the mentalhealth of patients, but also because noncompliance due to adverse eventsmay prevent materialization of the survival gain, which is derived onlyif the medication is taken as prescribed. Therefore, if mortality is expectedto decrease significantly, efforts to improve quality of life have to beescalated in proportion.
The hot flashes and vaginal dryness, with related dyspareunia, lossof libido, and loss of energy, seen in a large number of patients takingtamoxifen, are clearly antiestrogenic effects. Although never tested prospectively,there is a reasonable expectation that these tamoxifen-associated menopausalcomplaints could be ameliorated by hormone replacement therapy (HRT) inthe same fashion as are complaints associated with physiologic menopause.
According to many studies, the relative risk rates for the incidenceof breast cancer in populations of women without breast cancer are onlymarginally increased by HRT.[10,11] Because of an associated benefit onnonbreast cancer events, HRT use has been advocated for routine therapyof menopausal symptoms or as means to improve overall longevity.[12] Forsimilar reasons, the use of HRT in breast cancer survivors has been recentlyadvocated for amelioration of menopausal symptoms resulting from adjuvantchemotherapy or tamoxifen that affect quality of life.[13,20] The use ofHRT in breast cancer patients is particularly attractive if HRT is usedunder the protective umbrella of tamoxifen.[14] Preliminary evidence fromsmall phase II studies of HRT even without tamoxifen indicates that therate of breast cancer metastases may not be adversely affected by the useof HRT.[15-17] Furthermore, there are reasonable expectations that HRTuse may not only improve quality of life in patients who suffer from adversemenopausal symptoms, but HRT may add to other protective effects of tamoxifen,such as effects on lipids or prevention of bone loss as it does in patientswho do not have breast cancer.
On the negative side, thromboembolic episodes may be aggravated, andendometrial carcinogenesis remains a worrisome issue. Uterine cancer incidencecould be substantially reduced with the use of progesterone-containingHRT combinations due to the well-documented protective effect of pro- gestinson the estrogen-mediated increase of endometrial neoplasia.[18]
Because the addition of estrogen or progesterone may interfere withtamoxifen's effects on breast cancer, only very rigorous testing of theirvalue in controlled clinical trials will fully confirm their overall benefitin tamoxifen-treated breast cancer patients. At least one randomized trial[19]and several proposals for HRT trials in breast cancer patients are presentlyunder consideration by large centers or groups.[13,14] There has also beena call for studies to investigate the potential benefits of HRT treatmenton menopausal side effects or quality of life, and also possibly on overallsurvival, in breast cancer patients treated with or without tamoxifen.[13,19,20]
There is limited evidence for improvement of menopausal symptoms usingnonhormonal and unconventional approaches. The best studied agents arealpha-adrenergic agonists, such as clonidine and lofexidine. (Clonidineis used in tablet form or as a transdermal patch.) For some patients, betablockers such as propranolol have been used and anecdotal reports indicatea good outcome, although lightheadedness can be a problem. Tibolone isa steroid, with weak estrogenic, androgenic, and progestogenic activity,with no known adverse endometrial effects. Bellargal, a combination ofbelladonna alkaloids, ergotamine tartrate, and phenobarbital, is associatedwith improvement of hot flashes, although sedative effects, dry mouth,and vasoconstriction are possible side effects. The vitamin/herbal approachesinclude such agents as vitamin E, the Oriental or North American root ofginseng, primrose oil, and Astralagus.
Vitamin E is considered effective by some women, particularly as itmay relieve vaginal dryness. In most health food stores across North America,capsules or tea of ginseng are now available commercially. Ginseng hasbeen used in China for centuries as a general, nonspecific stimulant andaphrodisiac and is also considered to be good for many health conditions,including menopausal symptoms. Its effect is made more credible by thefact that ginseng root is a rich source of phytoestrogens.
Phytoestrogens, a class of phytosterols, are substances identified inplants that exert weak estrogenic effects with less well-documented stimulatorymitotic activity of pure estrogens. The higher intake of phytoestrogensin Asia is suspected to represent a common link to not only a lower breastcancer incidence in Asian countries, but also anecdotally with a reducedintensity of menopausal symptoms in women from underdeveloped countries.
Primrose is an edible North American plant, and its seeds contain primroseoil, a nontoxic oil containing gamma-linoleic acid, the source of its allegedactivity, which is considered to be of benefit in alleviating hot flashes.Astralagus, otherwise known as Huang Qi, is a root of a Chinese plant,one of the most popular components of Chinese herbal remedies consideredto have multiple health benefits including improvement of hot flashes.
Although these agents are used on a large scale in North America, dataon the efficacy of these nonconventional approaches are strictly limitedto anecdotal reports from individual women, greatly amplified by healthfood stores and holistic medicine lobbies. Thus, the issue of appropriateuse of these agents clearly needs clarification and testing in properlydesigned controlled studies.
Exercise and physical activity, perhaps by increasing serum and tissuelevels of steroids, androgens, or endorphins, as well as by lowering bloodpressure and reducing stress and headaches, may be an important step inrelieving many symptoms of menopause, including side effects of tamoxifen,and improving quality of life. However, the influence of exercise and thebiochemical processes mediating physical conditioning have been investigatedonly marginally, despite the fact that they have been frequently observedto be of major help in mediating subjective benefits on psychosocial trauma.
This report provides a review of tamoxifen's overall benefits and sideeffects, focusing on its impact on late mortality in breast cancer patientssurviving their primary disease and on its interaction with a patient'squality of life during treatment.
The discussion of tamoxifen's effect on late mortality focused on attemptsto distinguish between individual versus total effects. While the survivalimprovement for primary breast cancer has been firmly established, tamoxifen'sfull effect on late mortality in breast cancer survivors can be only estimatedfrom quantitative reports of past tamoxifen trials of adjuvant therapyfor breast cancer patients in which tamoxifen was also shown to affectthe incidence and mortality rates of other conditions.
The evaluation of tamoxifen's effect on total mortality is complex,as relevant conditions are affected at different times, by different ordersof magnitude, and frequently in opposite directions long after tamoxifenuse. What is emerging, however, is that only an age- matched evaluationintegrating all conditions affected by tamoxifen use will provide a completepicture of its real effect on overall mortality.
All side effects of tamoxifen, primarily those affecting quality oflife, require prospective studies because their management is emergingas one of the most important aspects of obtaining higher compliance fortamoxifen therapy. Maximum survival benefit would be expected in a compliantpopulation. Prospective testing of hormone replacement therapy is indicated,as evidence points toward enhanced quality of life and possibly improvedsurvival compared with tamoxifen alone. The nonconventional approaches,similar to the entry of any new agents or approaches in oncology, alsorequire controlled testing, on a scale perhaps even more rigorous thanconventional therapies as their use is already advocated or practiced virtuallywithout any attempt at outcome analysis.
Because their side effects are generally minimal, the nonconventionalapproaches may have the potential to substantially improve quality of liferelating to tamoxifen-associated or unassociated symptoms of menopause.They may thus contribute, in an unknown magnitude, toward a tamoxifen-mediatedincrease in late survival of breast cancer patients or populations of womenusing tamoxifen in prevention--a notion too important not to be testedprospectively.
1. Ragaz J, Coldman A: Age matched survival impact of tamoxifen in long-termbreast cancer survivors taking into analysis contralateral breast cancer,cardiovascular events, uterine cancer, and pulmonary emboli. Proc Am SocClin Oncol 14:112, 1995.
2. Jordan VC: The development of tamoxifen for breast cancer therapy,in Jordan VC (ed): Long-term Tamoxifen Treatment for Breast Cancer, pp3-27. Madison, University of Wisconsin Press, 1994.
3. Early Breast Cancer Trialists' Collaborative Group: Systemic treatmentof early breast cancer by hormonal, cytotoxic or immune therapy: 133 randomizedtrials involving 31,000 recurrences and 24,000 deaths among 75,000 women.Lancet 339:1-15, 1992.
4. Rutqvist LE, Johansson H, Signomklao T, et al: Adjuvant tamoxifentherapy for early stage breast cancer and second primary malignancies.Stockholm Breast Cancer Study Group. J Natl Cancer Inst 87:645-651, 1995.
5. McDonald CC, Stewart HJ: Fatal myocardial infarction in the ScottishAdjuvant Tamoxifen Trial. The Scottish Breast Cancer Committee. Br MedJ 303:435-437, 1991.
6. Rutqvist LE, Matteson A: Cardiac and thromboembolic morbidity amongpostmenopausal women with early-stage breast cancer in a randomized trialof adjuvant tamoxifen: the Stockholm Breast Cancer Study Group. J NatlCancer Inst 85:1398-1406, 1993.
7. Fisher B, Costantino JP, Redmond CK, et al: Endometrial cancer intamoxifen-treated breast cancer patients: findings from the National SurgicalAdjuvant Breast And Bowel Project (NSABP) B-14. J Natl Cancer Inst 86:527-537,1994.
8. Fornander T, Rutqvist LE, Cedermark B, et al: Adjuvant tamoxifenin early breast cancer: occurrence of new primary cancers. Lancet 1:117-120,1989.
9. Fisher B, Dignam J, Bryant J, et al: Five versus more than five yearsof tamoxifen therapy for breast cancer patients with negative lymph nodesand estrogen receptor-positive tumors. J Natl Cancer Inst 88:1529-1542,1996.
10. Steinberg KK, Thacker SB, Smith SJ, et al: A meta-analysis of theeffect of estrogen replacement therapy on the risk of breast cancer. JAMA265:1985-1990, 1991.
11. Colditz GA, Egan KM, Stampferer MJ, et al: Hormone replacement therapyand risk of breast cancer: results from epidemiologic studies. Am J ObstetGynecol 168:1473-1480, 1993.
12. Grady D, Rubin SM, Petitti DB, et al: Hormone therapy to preventdisease and prolong life in postmenopausal women. Ann Intern Med 117:1016-1037,1992.
13. Cobleigh MA, Berris RF, Bush T, et al: Estrogen replacement therapyin breast cancer survivors: A time for change. Breast Cancer Committeesof the Eastern Cooperative Oncology Group. JAMA 272:540-545, 1994.
14. Ragaz J: Hormone replacement therapy for tamoxifen-treated breastcancer patients suffering with postmenopausal symptoms. Proposal presentationto the Canadian National Cancer Institute, April, 1996.
15. Stoll BA, Parbhoo S: Treatment of menopausal symptoms in breastcancer patients. Lancet 1:1278-1279, 1988.
16. Di Saia PJ, Odicino F, Grosen EA, et al: Hormone replacement therapyin breast cancer. Lancet 342:1232, 1993.
17. Powles TJ, Hickish T, Casey S, et al: Hormone replacement afterbreast cancer. Lancet 342:60-61, 1993.
18. Gambrell RD Jr, Massey FM, Castaneda TA, et al: Use of progestogenchallenge test to reduce the risk of endometrial cancer. Obstet Gynecol55:732-728, 1980.
19. Vassilopoulou-Sellin R, Theriault RL: Randomized prospective trialof estrogen-replacement therapy in women with a history of breast cancer.Monogr Natl Cancer Inst 16:153-159, 1994.
20. Roy JA, Sawka CA, Pritchard KI: Hormone replacement therapy in womenwith breast cancer. Do the risks outweigh the benefits? J Clin Oncol 14:997-1006,1996.
21. Hunter M: The women's health questionnaire: a measure of mid-agedwomen's perceptions of their emotional and physical health. Psychol Health7:45-54, 1992.
22. Wiklund I, Holst J, Karlber J, et al: A new methodological approachto the evaluation of quality of life in postmenopausal women. Maturitas14:211-224, 1992.