Joy Melnikow, MD, MPH
In women without preexisting breast cancer, do tamoxifen, raloxifene, and tibolone reduce risk for breast cancer?
Included studies compared tamoxifen, raloxifene, or tibolone with placebo or each other for prevention of breast cancer in women without preexisting breast cancer, precursor conditions, or known susceptibility mutations. Outcomes were invasive and noninvasive breast cancer, venous thromboembolism, stroke, endometrial cancer, fractures, and cataracts.
MEDLINE, Cochrane Central Register of Controlled Trials, and Cochrane Database of Systematic Reviews (to Jan 2009); Web of Science; clinical trial registries; and reference lists were searched and drug manufacturers were consulted for double-blind randomized controlled trials (RCTs) that were published in English, treated ≥ 100 women for ≥ 3 months, and had invasive breast cancer as a primary or secondary outcome. 8 fair-to-good quality RCTs met the selection criteria: 4 placebo-controlled RCTs of tamoxifen (n = 28 193, median age 47 to 51 y), 2 placebo-controlled RCTs of raloxifene (n = 17 806, median age 67 to 68 y), 1 placebo-controlled RCT of tibolone (n = 4534, mean age 68 y), and 1 RCT comparing tamoxifen and raloxifene (n = 19 747, mean age 59 y). Mean duration of follow-up ranged from 3 to 13 years.
All 3 drugs reduced risk for invasive breast cancer (Table); for tamoxifen, the benefit continued for ≥ 3 to 5 years after treatment stopped (2 RCTs). No drug reduced risk for noninvasive breast cancer: risk ratio (RR) was 0.85 (CI 0.54 to 1.35) for tamoxifen and 1.5 (CI 0.75 to 2.9) for raloxifene. Risk for venous thromboembolism was increased with tamoxifen and raloxifene but not tibolone; in the direct-comparison trial, risk was lower with raloxifene than with tamoxifen (Table). Risk for stroke was increased with tibolone (RR 2.2, CI 1.1 to 4.2) but not the other 2 drugs. Risk for endometrial cancer was increased with tamoxifen (RR 2.1, CI 1.4 to 3.3) but not raloxifene (RR 1.1, CI 0.65 to 2.0). Tamoxifen reduced nonvertebral fractures (RR 0.66, CI 0.45 to 0.98), raloxifene reduced vertebral fractures (RR 0.61, CI 0.54 to 0.69), and tibolone reduced both types of fractures (RR 0.74, CI 0.58 to 0.93, and 0.55, CI 0.41 to 0.74, respectively). In the direct-comparison trial, risk for cataracts was lower with raloxifene than with tamoxifen (RR 0.79, CI 0.68 to 0.92), although neither drug differed from placebo.
In women without preexisting breast cancer, tamoxifen, raloxifene, and tibolone prevent invasive breast cancer and reduce risk for fractures but increase risk for thromboembolic events (tamoxifen and raloxifene), stroke (tibolone), and endometrial cancer (tamoxifen).
Medications to prevent primary breast cancer*
*Abbreviations defined in Glossary. RRR, RRI, NNT, NNH, and CI calculated from data in article using a random-effects model. NNT/NNH figures are per year. In the placebo groups, rates of invasive breast cancer ranged from 2.2 to 7.6/1000 woman-y, and rates of venous thromboembolism ranged from 0.31 to 2.5/1000 woman-y.
Melnikow J. Review: Tamoxifen, raloxifene, and tibolone prevent primary invasive breast cancer but increase risk for adverse outcomes. Ann Intern Med. ;152:JC3–4. doi: 10.7326/0003-4819-152-6-201003160-02004
Download citation file:
Published: Ann Intern Med. 2010;152(6):JC3-4.
Breast Cancer, Hematology/Oncology, Prevention/Screening.
Results provided by:
Copyright © 2019 American College of Physicians. All Rights Reserved.
Print ISSN: 0003-4819 | Online ISSN: 1539-3704
Conditions of Use