|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Originally published as JCO Early Release 10.1200/JCO.2002.06.029 on July 1 2002 © 2002 American Society for Clinical Oncology
American Society of Clinical Oncology Technology Assessment of Pharmacologic Interventions for Breast Cancer Risk Reduction Including Tamoxifen, Raloxifene, and Aromatase InhibitionByFrom the American Society of Clinical Oncology, Alexandria, VA. Address reprint requests to American Society of Clinical Oncology, Health Services Research Department, 1900 Duke St, Suite 200, Alexandria, VA 22314; email: guidelines{at}asco.orgThis article was published online ahead of print at www.jco.org.
OBJECTIVE: To update an evidence-based technology assessment of chemoprevention strategies for breast cancer risk reduction. POTENTIAL INTERVENTIONS: Tamoxifen, raloxifene, aromatase inhibition, and fenretinide. OUTCOMES: Outcomes of interest include breast cancer incidence, breast cancer–specific survival, overall survival, and net health benefit. EVIDENCE: A comprehensive, formal literature review was conducted for relevant topics. Testimony was collected from invited experts and interested parties. The American Society of Clinical Oncology (ASCO) prescribed technology assessment procedure was followed. VALUES: More weight was given to published randomized trials. BENEFITS/HARMS: A woman’s decision regarding breast cancer risk reduction strategies is complex and will depend on the importance and weight attributed to information regarding both cancer- and noncancer-related risks and benefits.
CONCLUSIONS: For women with a defined 5-year projected breast cancer risk of VALIDATION: The conclusions were endorsed by the ASCO Health Services Research Committee and the ASCO Board of Directors. SPONSOR: American Society of Clinical Oncology.
THE WORKING group (Appendix) reviewed the literature and, in a series of meetings and conference calls, considered recommendations from the original 1999 technology assessment.1 The technology assessment was limited to breast cancer chemoprevention. By this definition, surgical and lifestyle interventions were excluded from consideration but have been reviewed elsewhere.2,3
The technology assessment followed previously described procedures.1 A comprehensive literature review incorporated MEDLINE, Cancer Lit, PubMed, and scientific Internet sites (selected by link frequency). References were searched for the period from the last technology assessment (June 1999) through March 2002 using the following search terms: tamoxifen, raloxifene, aromatase inhibitors, retinoids and breast cancer, breast cancer risk reduction, and breast cancer risk communication. Of the 3,733 references identified and reviewed at least by title, 119 are referenced in the current report. Published information was updated by including key investigators in the Working Group and inviting corporate entities to provide emerging evidence (see Acknowledgment).
Conclusions of this technology assessment are outlined below (Table 1).
Tamoxifen For women with a 5-year projected breast cancer risk of  1.66%, tamoxifen (at 20 mg/d for 5 years) may be offered to reduce risk. Consideration of tamoxifen is appropriate for the goal of lowering the short-term risk of developing breast cancer. Risk/benefit models suggest that greatest clinical benefit with least side effects are derived from use of tamoxifen in younger (premenopausal) women (who are less likely to have thromboembolic sequelae and uterine cancer), women without a uterus, and women at higher breast cancer risk. Data do not as yet suggest that tamoxifen provides an overall health benefit or increases survival. Risk/benefit calculation for tamoxifen use is challenging. There is no simple scale to weigh the disparate clinical outcomes that vary in their morbidity and mortality risk. To inform potential tamoxifen users, the relative risk of outcomes under tamoxifen influence needs to be translated into absolute terms for each woman. In all circumstances, tamoxifen use should be discussed as part of an informed decision-making process with careful consideration of risks and benefits.
Raloxifene
Aromatase Inhibitor/Inactivators
Retinoids
Other Issues This technology assessment represents an ongoing process, and the Working Group will continue to monitor data and update recommendations in a timely manner. The basis for these recommendations follow. Clinical evidence is reviewed by chemopreventive agent.
TAMOXIFEN The absolute risk reduction in these trials was less than 2 per 100 women given tamoxifen for 5 years. The absolute risk reduction anticipated in an individual woman depends on her calculated breast cancer risk, with women at higher risk having greater potential benefit. For instance, the average 65-year-old woman with no family history has an anticipated risk reduction of 1 per 100, while a 50-year-old woman with two affected siblings and two prior biopsies but no germline mutation has an anticipated risk reduction of approximately 2.5 per 100. The initial report of the International Breast Cancer Intervention Study (IBIS-1) comparing tamoxifen with placebo and updates of the two other European chemoprevention tamoxifen trials7 substantially increase information regarding tamoxifen’s influence on breast cancer risk (Tables 2 and 3). Currently, 738 breast cancers have been reported in the four randomized tamoxifen trials, 54% more than available at the last technology assessment.
The IBIS-1 trial randomized 7,154 women at increased breast cancer risk to tamoxifen or placebo. The primary end point was breast cancer development (invasive and ductal carcinoma-in-situ) and the secondary end point was development of endometrial and other cancers. A relatively young population (median age, 50.8 years) at increased breast cancer risk that was identified using a newly developed risk assessment tool was randomized to tamoxifen (20 mg/d for 5 years) or placebo. Tamoxifen significantly decreased the risk of breast cancer development (odds ratio, 0.67; 95% CI, 0.49 to 0.91; P = .01), based on distribution of 169 breast cancers. Concurrent HRT use was permitted in the IBIS-1 study, was used by over 40% of participants, and had no detrimental effect on tamoxifen-associated breast cancer risk reduction. As in P-1, tamoxifen use in IBIS-1 reduced only receptor-positive breast cancers and had no influence on the receptor-negative cancers seen. A recent observational study reporting a tamoxifen-associated increase in receptor-negative breast cancers9 will require further evaluation. A meta-analysis7 now including the four randomized tamoxifen trials with updated results10 from all three European trials identified a 38% reduction in breast cancer with tamoxifen (odds ratio, 0.62; 95% CI, 0.42 to 0.89). These results support a significant influence of tamoxifen on reducing short-term breast cancer risk, with a magnitude of effect somewhat smaller than seen in the P-1 trial. Tamoxifen’s influence on survival in risk reduction trials. None of the four tamoxifen trials was designed to assess tamoxifen’s influence on survival. Available data do not yet suggest that tamoxifen favorably influences overall health or survival in the risk reduction setting. Currently, the number of deaths are closely comparable in the tamoxifen and control arms of the prevention trials.7 All three European trials continue blinded follow-up and will provide additional information on tamoxifen and breast cancer development, carry-over effects after tamoxifen discontinuation, and breast cancer–specific survival and overall survival in the future. Tamoxifen use with HRT. There is increasing experience with combined tamoxifen and HRT for breast cancer risk reduction. The Royal Marsden Hospital Trial included 523 women using both tamoxifen and HRT; subgroup analysis found no interaction on breast cancer development.5 In the Italian study,6 a subgroup analysis found tamoxifen significantly reduced breast cancer risk only in women receiving HRT, an analysis based on only nine cancers. The large IBIS-1 trial reported comparable tamoxifen-associated risk reduction in women with or without concurrent HRT use. In IBIS-1, combined HRT and tamoxifen was not associated with a greater short-term increase in endometrial cancer or vascular events compared with tamoxifen alone.7 Although no interaction between HRT and tamoxifen on breast cancer was identified, the association of HRT with increased breast cancer risk in observational studies11,12 and uncertainties regarding long-term effects of combined HRT and tamoxifen use argue against routine use of such a combination in a risk reduction setting. However, further evaluation of combined HRT and tamoxifen use remains an active area of investigation. Tamoxifen’s influence on contralateral breast cancer in adjuvant trials. In the last published Early Breast Cancer Trialists Cooperative Group (EBCTCG) analysis of adjuvant trials,13 a 47% reduction in relative risk of contralateral breast cancer was associated with 5 years of tamoxifen use (RR, 0.53; SD, 0.09; P < .00001), providing further support for a tamoxifen effect on new breast cancer development. The EBCTCG 2000 update identified 553 contralateral breast cancers and reported that contralateral breast cancer risk reduction was seen only in women who had an initial receptor-positive tumor.14 Supporting this observation is a recent report that women who develop receptor-negative breast cancer are likely to develop receptor-negative contralateral breast cancer.15 Tamoxifen duration for risk reduction. In the P-1 and IBIS trials, which provide the most evidence on risk reduction, the duration of tamoxifen use was 5 years.4,7 Indirect information on tamoxifen duration in this setting comes from follow-up of the NSABP B-14 trial in adjuvant disease, which found no additional benefit on contralateral breast cancer for continued tamoxifen use beyond 5 years.16 Tamoxifen’s effects on benign breast disease. A concern that tamoxifen’s effect on breast cancer risk may only reflect early therapy of invasive disease has been addressed by an analysis of benign breast disease in the P-1 trial. In this setting, tamoxifen was associated with a significant reduction in both the total number of biopsies and in the proportion of specimens from these biopsies showing atypical hyperplasia, typical hyperplasia, adenosis, cysts, and metaplasia.17 Tamoxifen’s effects in women with BRCA1/BRCA2 mutations. The NSABP has evaluated the effect of tamoxifen on breast cancer risk in P-1 participants with inherited BRCA1 or BRCA2 mutations.18 Of 288 women who developed breast cancers, 19 had either BRCA1 or BRCA2 mutations. Tamoxifen reduced breast cancer risk in women with BRCA2 mutations, in whom tumors were largely receptor-positive (RR, 0.38; 95% CI, 0.06 to 1.56), but not in women with BRCA1 mutations (RR, 1.67; 95% CI, 0.32 to 10.70), in whom tumors were more commonly receptor-negative. However, a matched case-control study reported significant protection against contralateral breast cancer in carriers of BRCA1 mutations treated with tamoxifen (odds ratio, 0.38; 95% CI, 0.19 to 0.74), while contralateral cancers in BRCA2 mutation carriers were less influenced (odds ratio, 0.63; 95% CI, 0.20 to 1.50).19 The extremely limited number of subjects in these reports and the mixed results seen preclude reliable assessment of tamoxifen effects in this setting. Given the overall number of identified mutation carriers, this important issue is unlikely to be resolved by further analyses of completed trials. Tamoxifen’s effects in women with prior breast disease. Women with a history of ductal carcinoma-in-situ20 or lobular carcinoma-in-situ are at recognized increased breast cancer risk and are reasonable candidates for tamoxifen use.4 Women with prior receptor-positive invasive breast cancer who have either not received tamoxifen or received tamoxifen for less than 5 years are also at increased breast cancer risk and can potentially benefit from tamoxifen.21,22 Tamoxifen in African-American women. Tamoxifen’s effectiveness in breast cancer risk reduction in African-American women has been inferred from a retrospective analyses involving 58 contralateral breast cancers seen in NSABP adjuvant trials,23 where a 52% reduction in relative risk of contralateral breast cancer was observed (RR, 0.48; 95% CI, 0.34 to 0.89). Other Positive and Negative Effects of Tamoxifen Use New information on tamoxifen has emerged from the NSABP P-1 study, the EBCTCG 2000 update, and meta-analyses of published literature.24,25 The effects of tamoxifen reflect more than 20 years of clinical trial use. Tamoxifen’s carry-over effects. The EBCTCG updated results from trials comparing 5 years of tamoxifen use to no therapy in more than 8,000 women with early breast cancer.14 Although not published, presented results informed the National Cancer Institute’s Adjuvant Breast Cancer Consensus Conference.26 In the EBCTCG update, time-dependent effects of tamoxifen use were identified. Five years of tamoxifen substantially reduced the risk for breast cancer recurrence and contralateral cancer that persisted for 5 to 10 years after termination of tamoxifen use. In addition, increased endometrial cancer risk was also persistent, supporting a previously identified trend.13,25,27 Publication of the EBCTG update is needed to fully evaluate these observations. Tamoxifen and quality of life, depression, and cognition. Tamoxifen use is associated with more frequent hot flashes and vaginal discharge.4 Development of cataracts also is more frequent (RR, 1.14; 95% CI, 1.01 to 1.29) with tamoxifen use, with the absolute risk increasing by 3 per 1,000 women.4 Tamoxifen did not adversely influence questionnaire-assessed depression risk,28 quality of life,29 or several other psychosocial or social functions.30,31 Other risk reduction trials have yet to report on these parameters. Tamoxifen’s effects on cognition are not resolved. A neuroimaging study compared levels of myoinositol (a glial marker) in women receiving tamoxifen, HRT, or no hormonal treatments. A significant time-dependent reduction in myoinositol was seen with tamoxifen, consistent with agonist action on brain predictive of a favorable neuroprotective effect.32,33 In a cross-sectional study of nursing home residents, women who reported past tamoxifen use had less frequent Alzheimer’s disease and greater activities of daily living compared with women who had never used tamoxifen.34 However, in an observational study using mailed questionnaires, current users complained of somewhat more memory problems,35 although no decrease in cognitive function was seen. Additional studies using sensitive methodologies that delineate memory and recall abilities are addressing this issue. Tamoxifen and cardiovascular events. In the NSABP P-1 trial, cardiovascular events, including myocardial infarctions, were similar in women receiving tamoxifen or placebo regardless of pre-existing coronary heart disease.36 Tamoxifen and vascular events. Vascular events include deep venous thrombosis, pulmonary embolism, stroke, and transient ischemic attack. A recent meta-analysis of published tamoxifen trials found the incidence of both venous thromboembolic events and strokes to be significantly greater in women receiving tamoxifen.24 The EBCTG update confirmed prior estimates of tamoxifen-associated excess mortality related to vascular events, largely pulmonary emboli of approximately one death per 1,000 postmenopausal women treated for 5 years.14 Since nearly half of thromboembolic events seen in the IBIS trial occurred within 3 months of surgery or fracture,7 tamoxifen use should be carefully re-evaluated in women using tamoxifen in a risk reduction setting after such events. Given common protocol exclusions for vascular events (Table 2) and identified risk, tamoxifen use for breast cancer risk reduction is relatively contraindicated and not recommended in women with a history of deep vein thrombosis, pulmonary embolus, stroke, or transient ischemic attack. Tamoxifen and endometrial cancer. Tamoxifen increases endometrial cancer risk in postmenopausal women with a uterus by approximately two- to four-fold.4,7 In the EBCTG update, tamoxifen-associated excess mortality related to endometrial cancer was approximately one death per 1,000 postmenopausal women with a uterus treated.14 New information on tamoxifen-related endometrial cancers comes from a case-control study with 309 endometrial cancers. Long-term tamoxifen users who developed endometrial cancer had a relatively unfavorable prognosis related to histology (more common malignant mixed mesodermal tumors or sarcomas of the endometrium) and higher stage.37 Endometrial cancer risk with tamoxifen was increased by prior estrogen use and obesity.38 Despite ongoing trials, use of progestins to mitigate tamoxifen’s effects on the endometrium, as widely used in HRT regimens, is not recommended given progestin’s activity as a breast mitogen.39,40 Prospective evaluation did not support the routine use of either endometrial biopsy41 or transvaginal ultrasound42 in ongoing monitoring for women with history of breast cancer receiving tamoxifen. Recommended follow-up for women receiving tamoxifen includes a yearly gynecologic examination and timely work-up of vaginal bleeding. Tamoxifen and gastrointestinal cancers. Although a meta-analysis of randomized trials identified increased gastrointestinal and colorectal malignancies associated with tamoxifen use,24 a nested case-control study found no increase in colorectal cancer among tamoxifen users,43 so this issue remains unsettled. Tamoxifen and fractures. Tamoxifen use was associated with a modest, nonsignificant reduction in fractures compared with placebo in the P-1 trial4 and with significantly fewer fractures compared with anastrozole in an adjuvant trial.44 Whether the latter represents a positive tamoxifen effect, a negative aromatase inhibitor effect, or some combination remains to be determined.
RALOXIFENE Data on raloxifene’s influence on breast cancer come almost exclusively from the Multiple Outcomes of Raloxifene Evaluation (MORE) study, in which 7,705 postmenopausal women with osteoporosis were randomized to raloxifene at 60 or 120 mg/d or placebo for 4 years and monitored for breast cancer development from the safety database. At the last technology assessment, a total of 54 breast cancers (39 invasive) were reported. The MORE study has been updated47-49 with the most recent report, including 77 verified breast cancers of which 59 were invasive.49 Raloxifene use (combining both dosage arms) continues to be associated with a significant reduction in relative risk of developing invasive breast cancers (RR, 0.28; 95% CI, 0.09 to 0.30), especially estrogen receptor–positive breast cancers. The absolute reduction in risk of developing breast cancer was 1.4 per 100 women given raloxifene for 5 years’ duration. The MORE study continues to follow women under an amended design. Reconsented participants have been continued on the original treatment assignment, with those on the 120-mg/d arm receiving a reduced dose of 60 mg/d (the FDA-approved osteoporosis dose). Clinical outcome will be evaluated at 6 and 8 years after entry. Cautions regarding interpretation of the raloxifene data expressed in the previous technology assessment are still applicable. A meta-analysis of several randomized trials with 10,575 women (including the MORE trial) previously reported a smaller (55%) relative reduction in breast cancer risk in the raloxifene arms.50 The current status of an updated report of breast cancer development in these raloxifene randomized trials is not available. In an intriguing subgroup analysis conducted in 7,290 of the MORE participants, the placebo group women with baseline estradiol levels greater than 10 pmol/L had a 6.8-fold higher rate of breast cancer than women with undetectable levels (P = .005 for trend).47 In addition, the raloxifene-associated reduction in breast cancer risk was largely in women with the highest baseline estradiol levels, which suggests that individualized stratification for breast cancer risk reduction interventions may be possible. This hypothesis will require prospective confirmation. Study of Tamoxifen and Raloxifene and Raloxifene for Use in the Heart: Definitive raloxifene outcome trials. Definitive assessment of raloxifene’s influence on breast cancer awaits the results from two ongoing randomized trials (Table 4). The Study of Tamoxifen and Raloxifene (STAR) trial has randomized more than 13,000 women of a 22,000 recruitment target and is scheduled to report outcome in 2008 or 2009.45 The Raloxifene for Use in the Heart (RUTH) study has completed recruitment with 10,011 postmenopausal women and has added breast cancer to cardiac disease as a second primary study end point.46 The RUTH trial, with baseline breast cancer risk assessment, periodic serial mammography, and randomization to either raloxifene 60 mg/d or placebo, will report breast cancer outcomes in 2005 after completion of the fourth-year mammograms.
Pending results of the outcome trials (Table 4), there is currently insufficient evidence to support routine use of raloxifene for breast cancer risk reduction. Raloxifene use in women with resected breast cancer. Raloxifene reduces mammographic breast density51 and breast cancer proliferative indices.52 However, as raloxifene has limited activity against advanced breast cancer when used after tamoxifen53 and does not decrease and may increase hot flash frequency,54 there is no basis for substituting raloxifene for tamoxifen in the treatment of adjuvant breast cancer patients experiencing menopausal symptoms. Currently, there are no published data regarding sequential tamoxifen and raloxifene use or use of raloxifene with HRT. Preclinical studies demonstrating raloxifene stimulation of tamoxifen-resistant breast cancers in athymic mice55 and raloxifene cross-resistance with tamoxifen53 suggest caution in raloxifene use in women with resected breast cancer who have completed tamoxifen adjuvant therapy. Other Positive and Negative Effects of Raloxifene Use Raloxifene’s toxicity profile remains largely as previously described.1 Published information on raloxifene is based on 4 years of use and is limited to postmenopausal women. Emerging reports regarding raloxifene effects are summarized below. Raloxifene and fracture risks. The FDA approved raloxifene 60 mg/d for osteoporosis prevention and therapy in postmenopausal women.56 The disproportionately greater decrease in vertebral fractures relative to bone density increase seen with raloxifene57 supports use of clinical outcomes rather than intermediate markers for definitive evaluation of interventions. Raloxifene and other effects. Raloxifene did reduce cardiovascular events when examined as a secondary end point in the previously mentioned MORE trial.58 Raloxifene is associated with a three-fold increase in potentially life-threatening vascular events (RR, 3.1; 95% CI, 1.5 to 6.2 for venous thrombolic events),59 comparable to the risk of these events with either tamoxifen or HRT use.60 The suggestion from preclinical and early clinical trials61 that raloxifene will have little influence on the endometrium awaits determination of endometrial cancer rates in ongoing longer-term randomized clinical trials. Raloxifene reduced uterine leiomyomas in one randomized trial62 and may reduce risk of pelvic floor surgery as well.63 Given the preclinical studies consistent with neuroprotection,64 raloxifene’s influence on cognition has received preliminary evaluation, with a trend toward less decline in tests of verbal memory and attention.65
AROMATASE INHIBITORS/INACTIVATORS There is interest in evaluating aromatase inhibitors for breast cancer risk reduction because estrogen levels are strongly related to higher breast cancer risk in postmenopausal women68 and estrogen levels are greatly reduced by aromatase inhibitors.66 The Anastrazole, Tamoxifen Alone and Combined trial. Relevant to breast cancer risk reduction are results from an adjuvant trial involving the aromatase inhibitor anastrozole (Arimidex; AstraZeneca). The Anastrazole, Tamoxifen alone and Combined (ATAC) trial44 is an international, multicenter, randomized, double-blind trial that randomized 9,366 patients with early-stage breast cancer to three treatment arms: tamoxifen 20 mg/d, anastrazole 1 mg/d, or the combination for 5 years. Primary study end points included disease-free survival and safety/tolerability; new (contralateral) breast cancer primary tumors were a secondary end point. Anastrazole, but not the anastrazole plus tamoxifen combination, was superior to tamoxifen in disease-free survival after a median follow-up of 33.4 months (discussed extensively by Winer et al69). In the ATAC trial, anastrazole use was also associated with a significant 58% reduction in new contralateral breast primary tumors (RR, 0.42 and 95% CI, 0.22 to 0.79 for anastrazole compared with tamoxifen groups; P < .0068). The absolute reduction in risk of developing a new contralateral breast cancer associated with anastrozole compared with tamoxifen was less than 1%. Currently, 62 new invasive contralateral breast cancers have occurred in the ATAC trial. In comparison, the tamoxifen-associated relative risk reduction of 49% in contralateral cancers reported in the NSABP B-14 adjuvant trial, which provided strong impetus for the P-1 trial, was based on the distribution of 83 contralateral cancers.70 In the ATAC trial, women on the anastrozole arm had significantly fewer vascular and endometrial events71 but more musculoskeletal events and fractures compared with those on the tamoxifen arm.44,71 Because there are no data on the long-term effects of aromatase inhibitor therapy, these agents should not be used in primary prevention for breast cancer outside of a research study. Proposed breast cancer risk reduction trials of aromatase inhibition. The effects of aromatase inhibitors (anastrozole, letrozole) and inactivators (exemestane) in advanced breast cancer66 and the emerging effects of anastrozole on contralateral breast cancer risk44 support further evaluation of anastrazole and other agents in this class for breast cancer risk reduction (Table 4).
RETINOIDS
EXOGENOUS ESTROGEN AND CHRONIC DISEASE RISK No full-scale randomized clinical trials of HRT have prospectively evaluated HRT’s influence on primary prevention of coronary heart disease or either coronary heart disease–associated mortality or all-cause mortality. More recent observational studies suggest an association with HRT use and an increase in breast cancer risk of approximately 30%,12 especially for longer use (> 5 years) and for regimens with progestins.11,39 Translated into absolute risk, approximately six excess breast cancers would be anticipated in 1,000 women using HRT for 10 years.75 Use of HRT is associated with an approximate three-fold increase in life-threatening vascular events and, when used without progestins, increased endometrial cancer risk.76 The approved indications for estrogens used as HRT include symptoms associated with moderate to severe menopause and prevention and management of osteoporosis. Although HRT is often proposed for improving overall health and reducing mortality, recent estimates of HRT’s influence on mortality are largely based on purported effects on cardiac events stemming from observational study results and lipid profile influence.77,78 The established effects of HRT in preventing osteoporosis have a relatively small impact on mortality because life-threatening fractures tend to occur later in life. A favorable effect of HRT on cognition has been predicted from preclinical and observational studies,79 but the data are somewhat mixed. A meta-analysis of observational studies identified a risk of dementia in estrogen users of 0.71 (95% CI, 0.53 to 0.96) compared with nonusers.65 However, a randomized trial of estrogen therapy for mild to moderate Alzheimer’s disease showed no effect of 1 year of therapy on disease progression or on global, cognitive, or functional outcomes.80 The Women’s Health Initiative Memory Study will address the question of HRT’s effect on cognitive function in a large population of healthy postmenopausal women.81 Observational findings on HRT and epithelial ovarian cancer are mixed, but a recent large case-control study associated users of HRT regimens with estrogen alone or sequentially added progestins with increased ovarian cancer risk.82 A full discussion of the risks and benefits of HRT is beyond the scope of the current technology assessment. Observational studies suggest an association with long-term HRT use and an approximate 35% reduction in primary cardiovascular events.76 However, randomized prospective clinical trials have recently evaluated HRT’s influence on clinical end points, including cardiac and vascular events. In these trials, HRT use did not result in secondary prevention of coronary vascular disease (CVD) or cerebral vascular events.83,84 In the large population (> 27,000) of generally healthy postmenopausal women randomized to HRT or placebo in the ongoing Women’s Health Initiative, interim analysis found an increase in the number of myocardial infarctions, strokes, and thromboembolic events in the initial years for women receiving HRT as compared with placebo.85 Continued follow-up of such trials will provide definitive assessment of HRT’s influence on CVD and overall mortality. Results from the Women’s Health Initiative are anticipated in 2005.86 Review of this emerging body of information has led several agencies, including the American Heart Association, to alter their recommendations regarding HRT. The Association’s recommendations for HRT and CVD now state that "firm recommendations for primary prevention await randomized clinical trial results and there is insufficient data to suggest that HRT should be initiated for the sole purpose of primary prevention of CVD."87 This technology assessment takes no position regarding HRT use in postmenopausal women. However, recommendation of HRT use for cardiovascular disease risk reduction and overall health or survival benefit should be approached using the same risk/benefit algorithm outlined for tamoxifen.
BREAST CANCER RISK ASSESSMENT The Gail model does not incorporate information needed to assess germline mutation carrier status or several other high-breast-cancer-risk situations, such as prior thoracic radiation.92 Thus, additional information must be collected to determine whether use of the Gail model is appropriate. Issues related to genetic susceptibility testing and, in particular, evaluation for BRCA1 and BRCA2 are reviewed elsewhere.78,88 Factors related to having 10% or greater risk of germline mutation include the following: known BRCA1 and BRCA2 family mutation, breast and ovarian cancer in the same family member, two or more family members under age 50 with breast cancer, male breast cancer, one or more family members under age 50 with breast cancer plus Ashkenazi ancestry, ovarian cancer plus Ashkenazi ancestry, and breast cancer before age 40. Women whose history suggests breast cancer germline mutation are not appropriate candidates for use of the Gail model, as their risk would be underestimated. Such women are candidates for genetic counseling referral. For women without prior breast cancer or who are not at increased risk for BRCA1/BRCA2, use of the Gail model is appropriate as long as there are no family members with breast cancer diagnosed before age 50. In those circumstances, the Claus model tables, which incorporate detailed family history, should be used.93 Information on breast cancer risk for both medical and nonmedical personnel is available from the Internet at the National Cancer Institute’s PDQ breast cancer prevention site, http://www.cancer.gov/cancer_information./ With regard to breast cancer risk assessment in the future, additional diagnostic and laboratory tests are being evaluated for breast cancer risk assessment. They include bone density, mammographic breast density, circulatory estradiol levels, and breast cells collected by a variety of techniques.94 The role of these procedures in clinical practice is beyond the scope of the present technology assessment.
COMPARISON OF QUANTITATIVE APPROACHES FOR TAMOXIFEN RISK/BENEFIT ESTIMATION
Gail Model to Estimate Positive and Negative Effects of Tamoxifen for Individuals The only model currently available in a format allowing calculation of an individual risk/benefit assessment for tamoxifen use was developed by Gail et al.97 Using information from a variety of sources, this model generates tabular estimates of net benefit and risk of tamoxifen use in groups of women according to 5-year breast cancer risk, age, presence or absence of uterus, and race. The number of health outcomes either induced or prevented by 5 years of tamoxifen use was estimated. A net benefit/risk index was derived by assigning weights to each clinical event and summing the benefits and harms. Because models make assumptions about potential benefit (greatest in women with high risk of breast cancer) and greatest risk (generally older women who are at more jeopardy for thromboembolism and uterine cancer), these models predict that young women should experience a better benefit/risk ratio. Such studies do not factor in other sequelae, such as menopausal symptoms or sexual dysfunction, that may be more problematic in younger women. Application of the Gail model suggests that tamoxifen is most beneficial in younger women at higher breast cancer risk and those without a uterus. African-American women are anticipated to have less tamoxifen benefit based largely on increased risk of a vascular event and lower fracture risk.98 Although clinically useful, the Gail model predicted outcomes only in the 5-year treatment period, the assumptions in developing the risk/benefit index have not been validated, and some results are sensitive to minor changes in weights assigned to utilities. The weights assigned to the clinical outcomes were neither empirically derived nor preference-based and do not reflect differences in mortality, morbidity, or timing of occurrence. In addition, the likelihood of incurring most adverse effects except for endometrial cancer was based only on age not individual disease risk. Despite these concerns, the Gail model provides the best available beginning framework for risk calculation. New strategies and decision aids that incorporate mortality estimates and carry-over effects of tamoxifen are under development99 and are described below. Other Models to Estimate Positive and Negative Effects of Tamoxifen Smith and Hillner,100 in a cost-effectiveness analysis based on tamoxifen effects seen in P-1 over 5 years, found tamoxifen use to be cost-effective overall, considering all medical event–related costs compared with no tamoxifen use. Duffy and Nixon101 used meta-analytic techniques to estimate the impact of tamoxifen among BRCA1/BRCA2 carriers, finding tamoxifen to be more effective among BRCA2 carriers (27% risk reduction) than among BRCA1 carriers (no significant reduction). The impact of tamoxifen on BRCA1/BRCA2 carriers was inferred by assuming that the effect of tamoxifen is dependent on estrogen receptor status, determining the proportion of estrogen receptor–positive tumors among BRCA1 and BRCA2 mutation carriers, and computing a weighted average for BRCA1/BRCA2 carriers. Markov Modeling of Clinical Outcomes In Markov modeling, each clinical outcome is valued according to its survival or quality-adjusted survival impact, with time-dependent rates and utilities assigned to each outcome.102 Survival projections involve extrapolation based on assumptions concerning long-term effects of treatment (where information is often limited). Differences between the various published Markov models exploring tamoxifen impact are outlined below. The models of Hershman et al103 and Grann et al104 predict that tamoxifen use increases life expectancy of higher-risk women by 69 days for 35-year-olds and 27 days for 60-year-olds. The practical application of the predictions of quality-adjusted survival in this model is limited by the assignment of utilities (hip fracture was assigned a lower utility than metastatic breast cancer). Women with and without a uterus were treated equally, precluding determinations involving endometrial cancer risk. Will et al,105 using data from Canadian cancer registries and vital statistics to simulate effects of tamoxifen under a range of assumptions, incorporated all P-1 outcomes and calculated that tamoxifen would extend life expectancy among women whose 5-year predicted risk was 3.32% or greater. Col et al99 used a Markov model to address the impact of tamoxifen among postmenopausal women at varying levels of risk for breast and endometrial cancer and hip fracture. This model incorporates individual risk factors for both the benefits and principal harms of tamoxifen and is the only published model to include residual impact of tamoxifen based on the most recent EBCTCG update.14 This model predicts that tamoxifen use in 50-year-old women without a uterus results in an increased life expectancy of 1 to 4 months, whereas women with a uterus have gains only if they are at higher breast cancer risk. Tamoxifen carry-over effects on breast and endometrial cancer risk have a strong impact on the benefit/risk survival profile. These Markov models are complex and are constrained by the quality of the informing data. Despite the limitations, each model adds insight into assessing the risks and benefits of tamoxifen. None of the Markov models described has been assessed for validity or is available in a format allowing for individual patient data entry. Attempts are underway to develop such programs for routine clinical use.
CHALLENGES IN COMMUNICATING TAMOXIFEN’S RISKS AND BENEFITS Women typically overestimate their risk of breast cancer,106 emphasizing the importance of effective communication of breast cancer risk in this setting. Relative risk describes the ratio of the risk of disease in one group compared with that in another, does not take into consideration a person’s baseline risk, and does not describe the magnitude of the absolute risk. Absolute risk varies according to baseline level of risk and could be very small when the disease is uncommon. Overall, women should be given information that presents the risks and benefits of any intervention using both absolute and relative terms.107 How risk information is presented,108-110 worded,107 and framed111,112 may affect its interpretation.113 Framing the benefits (ie, gains) of treatment in relative rather than absolute terms can affect a patient’s perception of a therapy’s effectiveness,114 making the benefits of a treatment appear more favorable115 or, conversely, emphasizing its risks (ie, losses).116 Most framing studies have focused on a treatment’s effect on a single outcome over a single time horizon.112 Women deciding on tamoxifen therapy need to consider its effects (both beneficial and harmful) on several outcomes over an extended time horizon. A study examining patient preferences for communicating complex risk information117 found that patients preferred risk estimates framed in absolute rather than relative terms, graphical and textual explanations provided together, and risk estimates given over more than one time horizon. In addition, use of contextual and graphical displays may improve understanding of numerical risk/benefit information.110 Efforts are underway to develop such platforms for general clinical use. In sum, the communication of tamoxifen’s risks and benefits should include both absolute and relative information over a relevant time period. Attention should be paid to how the information is framed and presented.
CONCORDANCE WITH OTHER AGENCY RECOMMENDATIONS
The appendix listing the members of the ASCO Risk Reduction Update Working Group is available online at www.jco.org. 
The Working Group expresses its gratitude to Robert F. Ozols, MD, Trevor J. Powles, MD, and the American Society of Clinical Oncology’s Health Services Research Committee for their thoughtful review of earlier drafts of the technology assessment.
1. Chlebowski RT, Collyar DE, Somerfield MR, et al: American Society of Clinical Oncology technology assessment on breast cancer risk reduction strategies: Tamoxifen and raloxifene. J Clin Oncol 17: 1939-1955, 1999
2. Chlebowski R: Primary care: Reducing the risk of breast cancer. N Engl J Med 343: 191-198, 2000 3. Chlebowski RT: Breast cancer risk reduction: Strategies for women at increased risk. Annu Rev Med 51: 519-540, 2002[CrossRef]
4. Fisher B, Costantino JP, Wickerham DL, et al: Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 study. J Natl Cancer Inst 90: 1371-1388, 1998 5. Powles T, Eeles R, Ashley S, et al: Interim analysis of the incidence of breast cancer in the Royal Marsden Hospital Tamoxifen Randomized Chemoprevention Trial. Lancet 352: 98-101, 1998[Medline] 6. Veronesi U, Maisonneuve P, Costa A, et al: Prevention of breast cancer with tamoxifen: Preliminary findings from the Italian randomized trial among hysterectomized women. Lancet 352: 93-97, 1998[Medline] 7. Cuzick J: Update on new studies in Europe. Eur J Cancer 38: S20, 2002 (abstr) 8. Cuzick J: A brief review of the International Breast Cancer Intervention Study (IBIS), the other current breast cancer prevention trials, and proposals for future trials. Ann NY Acad Sci 949: 123-133, 2001[Medline]
9. Li CI, Malone KE, Weiss NS, et al: Tamoxifen therapy for primary breast cancer and risk of contralateral breast cancer. J Natl Cancer Inst 93: 1008-1013, 2001 10. Veronesi U, Maisonneuve P, Sacchini V, et al: Tamoxifen for breast cancer among hysterectomised women. Lancet 359: 1122-1124, 2002[CrossRef][Medline]
11. Ross RK, Paganini-Hill A, Wan PC, et al: Effect of hormone replacement therapy on breast cancer risk: Estrogen versus estrogen plus progestin. J Natl Cancer Inst 92: 328-332, 2000
12. Chen CL, Weiss NS, Newcomb P, et al: Hormone replacement therapy in relation to breast cancer. JAMA 287: 734-741, 2002 13. Early Breast Cancer Trialists’ Collaborative Group: Tamoxifen for early breast cancer: An overview of the randomised trials. Lancet 351: 1451-1467, 1998[CrossRef][Medline] 14. Peto R: Tamoxifen as Adjuvant Breast Cancer Therapy. Presented at the NIH Consensus Development Conference on Adjuvant Therapy for Breast Cancer, http://videocast.nih.gov, 2000 15. Swain SM, Wilson J, Eleftherios M, et al: Estrogen receptor (ER) status of primary breast cancer is predictive of ER status of contralateral breast cancer (CBC). Proc Am Soc Clin Oncol 21: 38a, 2002 (abstr 150)
16. Fisher B, Dignam J, Bryant J, et al: Five versus more than five years of tamoxifen for lymph node-negative breast cancer: Updated findings from the National Surgical Adjuvant Breast and Bowel Project B-14 randomized trial. J Natl Cancer Inst 93: 684-690, 2001 17. Tan-Chiu E, Costantino J, Wang J, et al: The effect of tamoxifen on benign breast disease: Findings from the Nationial Surgical Adjuvant Breast and Bowel Project (NSABP) Breast Cancer Prevention Trial (BCPT). San Antonio Breast Cancer Symposium, 2001 (abstr 7)
18. King MC, Wieand S, Hale K, et al: Tamoxifen and breast cancer incidence among women with inherited mutations in BRCA1 and BRCA2: National Surgical Adjuvant Breast and Bowel Project (NSABP-P1) Breast Cancer Prevention Trial. JAMA 286: 2251-2256, 2001 19. Narod SA, Brunet JS, Ghadirian P, et al: Tamoxifen and risk of contralateral breast cancer in BRCA1 and BRCA2 mutation carriers: A case-control study—Hereditary Breast Cancer Clinical Study Group. Lancet 356: 1876-1881, 2000[CrossRef][Medline] 20. Fisher B, Dignam J, Wolmark N, et al: Tamoxifen in treatment of intraductal breast cancer: National Surgical Adjuvant Breast and Bowel Project B-24 randomised controlled trial. Lancet 353: 1993-2000, 1999[CrossRef][Medline]
21. Delozier T, Switsers O, Genot JY, et al: Delayed adjuvant tamoxifen: Ten-year results of a collaborative randomized controlled trial in early breast cancer (TAM-02 trial). Ann Oncol 11: 515-519, 2000 22. Gradishar WJ, Hellmund R: A rationale for the reinitiation of adjuvant tamoxifen therapy in women receiving fewer than 5 years of therapy. Clin Breast Cancer 2: 282-286, 2002[Medline] 23. McCaskill-Stevens W, Bryant J, Costantino J, et al: Incidence of contralateral breast cancer (CBC), endometrial cancer (EC) and thromboembolic events (TE) in African American (AA) women receiving tamoxifen for treatment of primary breast cancer. Proc Am Soc Clin Oncol 19: 70a, 2000 (abstr 269) 24. Braithwaite R, Law J, Chlebowski RT, et al: A meta-analysis of vascular and neoplastic events associated with tamoxifen. Med Decision Making 20: 485, 2000 (abstr)
25. Mourits MJ, De Vries EG, Willemse PH, et al: Tamoxifen treatment and gynecologic side effects: A review. Obstet Gynecol 97: 855-866, 2001 26. National Institutes of Health: National Institutes of Health consensus development conference statement: Adjuvant therapy for breast cancer. J Natl Cancer Inst Monogr 30: 5-15, 2001 27. Bertelli G, Valenzano M, Del Mastro L, et al: Endometrial abnormalities in breast cancer patients receiving adjuvant tamoxifen persist after treatment completion. Proc Am Soc Clin Oncol 20: 36a, 2001 (abstr 142)
28. Day R, Ganz PA, Costantino JP: Tamoxifen and depression: More evidence from the National Surgical Adjuvant Breast and Bowel Project’s breast cancer prevention (P-1) randomized study. J Natl Cancer Inst 93: 1615-1623, 2001
29. Day R, Ganz PA, Constantino JP, et al: Health-related quality of life and tamoxifen in breast cancer prevention: A report for the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Clin Oncol 17: 2659-2669, 1999
30. Berglund G, Nystedt M, Bolund C, et al: Effect of endocrine treatment on sexuality in premenopausal breast cancer patients: A prospective randomized study. J Clin Oncol 19: 2788-2796, 2001
31. Fallowfield L, Fleissig A, Edwards R, et al: Tamoxifen for the prevention of breast cancer: Psychosocial impact on women participating in two randomized controlled trials. J Clin Oncol 19: 1885-1892, 2001 32. Chlebowski RT, Ernst T, Chang L, et al: Tamoxifen and estrogen effects on brain chemistry determined by MRI spectroscopy. Proc Am Soc Clin Oncol 20: 28a, 2001 (abstr 108)
33. Ernst T, Chang L, Cooray D, et al: The effects of tamoxifen and estrogen on brain metabolism in elderly women. J Natl Cancer Inst 94: 592-597, 2002 34. Breuer B, Anderson R: The relationship of tamoxifen with dementia, depression, and dependence in activities of daily living in elderly nursing home residents. Women Health 31: 71-85, 2000[CrossRef][Medline] 35. Paganini-Hill A, Clark LJ: Preliminary assessment of cognitive function in breast cancer patients treated with tamoxifen. Breast Cancer Res Treat 64: 165-176, 2001 36. Reis SE, Costantino JP, Wickerham DL, et al: Cardiovascular effects of tamoxifen in women with and without heart disease: Breast cancer prevention trial. J Natl Cancer Inst 93: 17-21, 2001 37. Bergman L, Beelen ML, Gallee MP, et al: Risk and prognosis of endometrial cancer after tamoxifen for breast cancer: Comprehensive Cancer Centres’ ALERT Group—Assessment of Liver and Endometrial Cancer Risk Following Tamoxifen. Lancet 356: 881-887, 2000[CrossRef][Medline]
38. Bernstein L, Deapen D, Cerhan JR, et al: Tamoxifen therapy for breast cancer and endometrial cancer risk. J Natl Cancer Inst 91: 1654-1662, 1999
39. Schairer C, Lubin J, Troisi R, et al: Menopausal estrogen and estrogen-progestin replacement therapy and breast cancer risk. JAMA 283: 485-491, 2000 40. McCarty MF: Androgenic progestins amplify the breast cancer risk associated with hormone replacement therapy by boosting IGF-I activity. Med Hypotheses 56: 213-216, 2001[CrossRef][Medline]
41. Barakat R, Gilewski T, Almadrones L, et al: Effect of adjuvant tamoxifen on the endometrium in women with breast cancer: A prospective study using office endometrial biopsy. J Clin Oncol 18: 3459-3463, 2000
42. Gerber B, Krause A, Mnller H, et al: Effects of adjuvant tamoxifen on the endometrium in postmenopausal women with breast cancer: A prospective long-term study using transvaginal ultrasound. J Clin Oncol 18: 3464-3470, 2000 43. Cook LS, Weiss NS, Pharris-Ciurej N, et al: Colorectal cancer following tamoxifen therapy for breast cancer (United States). Cancer Causes Control 12: 405-410, 2001[CrossRef][Medline] 44. Baum M on behalf of the ATAC Trialist Group: The ATAC (Arimidex, Tamoxifen Alone or in Combination) adjuvant breast cancer trial in postmenopausal women. Breast Can Res Treat 21: 210, 2001 (abstr 8) 45. Vogel VG: Follow-up of the breast cancer prevention trial and the future of breast cancer prevention efforts. Clin Cancer Res 7:4413s-4418s; discussion 4411s-4412s, 2001 46. Mosca L, Barrett-Connor E, Wenger NK, et al: Design and methods of the Raloxifene Use for the Heart (RUTH) study. Am J Cardiol 88: 392-395, 2001[CrossRef][Medline]
47. Cummings SR, Duong T, Kenyon E, et al: Serum estradiol level and risk of breast cancer during treatment with raloxifene. JAMA 287: 216-220, 2002
48. Lippman ME, Krueger KA, Eckert S, et al: Indicators of lifetime estrogen exposure: Effect on breast cancer incidence and interaction with raloxifene therapy in the multiple outcomes of raloxifene evaluation study participants. J Clin Oncol 19: 3111-3116, 2001 49. Cauley JA, Norton L, Lippman ME, et al: Continued breast cancer risk reduction in postmenopausal women treated with raloxifene: 4-year results from the MORE trial—Multiple Outcomes of Raloxifene Evaluation. Breast Cancer Res Treat 65: 125-134, 2001[CrossRef][Medline]
50. Jordan VC, Glusman JE, Eckert S, et al: Incident primary breast cancers are reduced by raloxifene: Integrated data form multicenter, double blind, randomized trials in
51. Freedman M, Martin JS, O’Gorman J, et al: Digitized mammography: A clinical trial of postmenopausal women randomly assigned to receive raloxifene, estrogen, or placebo. J Natl Cancer Inst 93: 51-56, 2001
52. Dowsett M, Bundred NJ, Decensi A, et al: Effect of raloxifene on breast cancer cell Ki67 and apoptosis: A double-blind placebo-controlled, randomized clinical trial in postmenopausal patients. Cancer Epidemiol Biomarkers Prev 10: 961-966, 2001 53. Gradishar W, Glusman J, Lu Y, et al: Effects of high dose raloxifene in selected patients with advanced breast carcinoma. Cancer 88: 2047-2053, 2000[CrossRef][Medline] 54. Cohen FJ, Lu Y: Characterization of hot flashes reported by healthy postmenopausal women receiving raloxifene or placebo during osteoporosis prevention trials. Maturitas 34: 65-73, 2000[CrossRef][Medline]
55. O’Regan RM, Gajdos C, Dardes RC, et al: Effects of raloxifene after tamoxifen on breast and endometrial tumor growth in athymic mice. J Natl Cancer Inst 94: 274-283, 2002
56. Johnston CCJ, Bjarnason NH, Cohen FJ, et al: Long-term effects of raloxifene on bone mineral density, bone turnover, and serum lipid levels in early postmenopausal women: Three-year data from 2 double-blind, randomized, placebo-controlled trials. Arch Intern Med 160: 3444-3450, 2000 57. Sarkar S, Mitlak BH, Wong M, et al: Relationships between bone mineral density and incident vertebral fracture risk with raloxifene therapy. J Bone Miner Res 17: 1-10, 2002[CrossRef][Medline]
58. Barrett-Connor E, Grady D, Sashegyi A, et al: Raloxifene and cardiovascular events in osteoporotic postmenopausal women: Four-year results from the MORE (Multiple Outcomes of Raloxifene Evaluation) randomized trial. JAMA 287: 847-857, 2002
59. Cummings SR, Eckert S, Krueger KA, et al: The effect of raloxifene on risk of breast cancer in postmenopausal women: Results from the MORE randomized trial—Multiple Outcomes of Raloxifene Evaluation. JAMA 281: 2189-2197, 1999
60. Khovidhunkit W, Shoback DM: Clinical effects of raloxifene hydrochloride in women. Ann Intern Med 130: 431-439, 1999 61. Cohen FJ, Watts S, Shah A, et al: Uterine effects of 3-year raloxifene therapy in postmenopausal women younger than age 60. Obstet Gynecol 95: 104-110, 2000 |
TOP
ABSTRACT
INTRODUCTION
12,000 postmenopausal women. Proc Am Soc Clin Oncol 17: 122a, 1998 (abstr 466)