Tamoxifen

IBIS Investigators First results from the International Breast Cancer Intervention Study (IBIS-I): a randomized prevention trial. Lancet 2002; 360:817 24.

* BACKGROUND Results have been mixed from 3 previous clinical trials of the use of tamoxifen for preventing breast cancer in high-risk women. A recently updated analysis from 1 negative study showed a decrease in the incidence of estrogen-receptor-positive breast cancer, adding support for the effectiveness of tamoxifen in some high-risk women. Tamoxifen therapy, however, is associated with significant adverse outcomes that may outweigh any benefits.

* POPULATION STUDIED Recruitment took place among outpatients at breast screening centers, primary care offices, and family history clinics. The media and contact of relatives of women with breast cancer provided additional recruitment. Women aged 35 to 70 years met criteria for entry if their risk for developing breast cancer was at least double that for women between 45 and 70 years, at least 4 fold that for women between 40 and 44 years, and 10-fold that for women aged 35 to 39 years. Risk was determined by an unpublished alternative to the Gail model. Patients were excluded if they had current or desired pregnancy, previous invasive cancer, previous deep vein thrombosis or pulmonary embolism, current use of anticoagulant agents, or a life expectancy of less than 10 years.

* STUDY DESIGN AND VALIDITY In this randomized, placebo-controlled, double-blinded trial, eligible women were randomized to receive either 20 mg tamoxifen (N=3573) or a matched placebo (N=3566) daily for 5 years. Thirteen women with findings on mammogram at the time of randomization were excluded from subsequent analysis. Participants received clinical evaluations every 6 months during the 5-year active treatment interval. Follow-up continued for up to 5 years beyond treatment by annual questionnaire or clinical visit. Symptoms, side effects, diagnoses, procedures, and concomitant medications were recorded at each evaluation. A mammogram was performed every 12 to 18 months. External reviewers, masked to treatment allocation, evaluated all end points and deaths.

Several features of the study design ensured internal validity. Randomization occurred by contact with a central office and was stratified by study center. Initial blocks of 8 were randomly distributed into blocks of 6 to 10 to ensure full masking. Analysis was performed on an intention-to-treat basis. Median follow-up was 50 months and the authors estimated that full compliance to 5 years was 64% in the tamoxifen group and 74% in the placebo group, although actual numbers of patients lost to follow-up were not provided. At the end of the data collection period, only 25% of women had completed a full 5 years of treatment (26.9% placebo vs 23.4% tamoxifen).

* OUTCOMES MEASURED The primary outcome measure was the frequency of breast cancer including ductal carcinoma in situ. Secondary outcomes included side effects, thromboembolic events, procedures, and deaths.

* RESULTS Women given tamoxifen had fewer breast cancers (n=60, 1.9%) than individuals given placebo (n=101, 2.8%; numbers needed to treat=112 for 5 years of treatment). As expected, prevention occurred at a cost of a greater incidence of venous thromboembolic events (1.2% vs 0.5%; numbers needed to harm [NNH]=137 for 5 years of treatment) and symptoms such as hot flashes and breast tenderness, and procedures including dilation and curettage and hysterectomy. More importantly, significantly more women on tamoxifen (n=25, 0.7%) died as compared with those on placebo (n=11, 0.3%; NNH=256 for 5 years of treatment).

RECOMMENDATIONS FOR CLINICAL PRACTICE

In women at high risk of breast cancer, tamoxifen is effective in reducing the incidence of the disease. The intervention, however, is associated with significant risk. During the 5-year period of this study, although the number of breast cancers was reduced, the number of serious adverse effects and deaths were higher in the treated group. Women at a lower risk of breast cancer than those studied would be even less likely to benefit from tamoxifen while risking the same serious adverse outcomes. For the small proportion of women with a high risk of breast cancer and a low risk of adverse events, discussion of this intervention may be warranted.

COPYRIGHT 2002 Dowden Health Media, Inc.
COPYRIGHT 2002 Gale Group