A systematic review of pharmacological primary prevention of breast cancer carried out for the US Agency for Healthcare Research and Quality found that tamoxifen, raloxifene, and tibolone do reduce the risk, but that they all also cause significant adverse effects.

Two drugs are licensed in the US for use as prophylaxis against primary breast cancer in women at high risk, tamoxifen and raloxifene; tibolone has also been studied for this use. There is limited evidence on the comparative efficacy of the drugs, or on the major adverse effects associated with their use. This comparative effectiveness review was commissioned by the AHQR to clarify the drugs’ potential place in therapy.

The authors carried out a comprehensive literature search for randomised controlled trials (RCT) published in English, and comparing the drugs with placebo or head to head as primary prophylaxis in women at higher risk of breast cancer; minimum study duration was 3 months and minimum study size 100 participants. For assessment of safety, non-randomised studies were also included.

The initial literature search identified 4,230 articles, of which 352 were retrieved for full assessment. After full text assessment, 58 were selected as relevant to the review. There were 8 RCT (4 tamoxifen vs. placebo, 2 raloxifene vs. placebo, 1 tamoxifen vs. raloxifene, and 1 tibolone vs. placebo): there were significant variations in participant age and study duration, and most were of moderate to good quality.

 
In placebo-controlled trials, all three drugs reduced the incidence of invasive breast cancer in women in midlife and older women by approximately 30% to 68%: tamoxifen risk ratio 0.70 (95% CI, 0.59 to 0.82; 4 trials), raloxifene RR 0.44 (95% CI, 0.27 to 0.71; 2 trials), and tibolone RR 0.32 (95% CI, 0.13 to 0.80; 1 trial). In two tamoxifen trials that provided post-treatment follow-up, the risk reduction continued for at least 3 to 5 years. The absolute risk reduction is about 7 to 10 events /1000 women per year based on five years use. All drugs significantly reduced risk of fractures.

In the harms assessment, tamoxifen and raloxifene increased the risk of thromboembolic events by about 4 to 7 events/1000 women per year, raloxifene causing fewer events than tamoxifen: tamoxifen RR 1.93 (95% CI, 1.41 to 2.64; 4 trials), raloxifene RR 1.60 95% CI, 1.15 to 2.23; 2 trials). In the trials providing post-treatment data, the risk fell to normal after discontinuation. Tamoxifen is associated with an increase in risk of endometrial cancer compared to placebo by 4 events/1000 women per year (RR, 2.13; 95% CI, 1.36 to 3.32; 3 trials). Based on limited data, tibolone does not increase the risk for thromboembolic events but increases the risk of stroke in older women (age 70+).

The authors conclude that these three drugs decrease the risk for breast cancer, but increase the risk of other adverse events. They note that biases, trial heterogeneity, and the lack of head-to-head trials limit their review. There are also gaps in the evidence, including the optimum doses, duration, and timing of the medications; data on long-term effects are limited, and there is limited evidence covering non-white and premenopausal women. They note also that clinical trial populations tend to be healthier than the general population, and because of study heterogeneity the results should be applied with caution.