Epidemiological and experimental evidence indicates that oestrogens are involved in the carcinogenic promotion of human breast cancer. We have undertaken a pilot trial of tamoxifen, an anti-oestrogen, compared to placebo given to 200 women at a high risk of developing breast cancer. The results of this trial show that acute toxicity is low and that accrual and compliance are satisfactory. Furthermore, biochemical monitoring of lipids and clotting factors indicate that tamoxifen may reduce the risk of cardiovascular deaths. At this stage no untoward long-term risks have been identified, and it is therefore proposed that a large multicentre trial should be started.