Epidemiological and experimental evidence indicate that oestrogenic activity plays an important role in the promotion of human breast cancer. This raises the possibility that anti-oestrogenic intervention could prevent the development of this disease.In order to detect a 25% reduction in incidence of breast cancer between a treatment and control population, at least 300 breast cancers would need to develop. With a high-risk group of women, such as those with a family history, between 40 and 60 years old, 10000 women would be required with a 10-year follow-up for 250-300 cancer to develop. This would indicate that to have a reasonable chance of detecting a significant prevention of breast cancer, by an anti-oestrogenic intervention, 5 000 treatment and 5 000 control women with a high risk of developing breast cancer would be needed. However, before such a major trial could be attempted it was essential to evaluate the ethics, logistics, patient and doctor acceptability, acute toxicity, patient accrual and compliance of tamoxifen, in a prevention context, in a small feasibility trial. We have therefore started a double-blind placebo controlled feasibility trial designed to accrual 200 women aged between 35 and 65 with a family history of breast cancer. Between October 1986 and July 1987 a total of 124 patients were randomised to receive either tamoxifen or placebo. With these patients as background the present paper will outline the various problems, including acute toxicity, compliance and patient acceptability for tamoxifen versus placebo in a prevention trial.Key words: Breast cancer, chemoprevention, tamoxifen, randomised trial.Experimental evidence indicates that oestrogen is involved in carcinogenesis of mouse mammary tumours. Both the DMBA and NMU induced mouse mammary tumours require intact ovarian function in order to develop. Ovariectomy will prevent tumour development, but normal tumour growth is restored by oestrogen administration even after a delay of several weeks. Ovarian-stimulated or oestrogen-stimulated tumour development is prevented by concomitant tamoxifen administration ( I , 2). These results indicate that chemical induction requires oestrogenic promotion in order to complete the carcinogenic process.Epidemiological evidence would seem to support the hypothesis that ovarian function is required to promote naturally induced breast cancer by background radiation, viruses or chemical induction carcinogens (3). For example, early menarche or naturally occurring late menopause are all associated with an increased risk of breast cancer (4). Conversely, the incidence of breast cancer may be substantially reduced by oophorectomy or radiation menopause (5, 6), the degree of protection related directly to the age of artificial menopause (6).Although ovarian function seems necessary for endocrine tumour promotion, it is not clear which hormones are involved. There is little or no evidence that raised oestrogen level is a factor, but it is possible that increased oestrogenic activity on breast ...
