In the Royal Marsden Hospital tamoxifen prevention study, 2500 women at increased risk of developing breast cancer because of family history of the disease were randomised to receive tamoxifen 20 mg daily or placebo for 8 years. 70 woman developed primary breast cancer; 36 whilst on placebo, 34 on tamoxifen. Family history out to at least 2nd degree relatives was taken from all women in the study. DNA from peripheral blood from 67 of the 70 women was analysed for coding mutations in the BRCA1 and BRCA2 genes by CSGE analysis of the entire coding region of both genes. 7 mutations were found, 2 in BRCA1 and 5 in BRCA2; 4 would be expected to be pathogenic as these were nonsense/frameshifts. 3 were rare variants which were not present in 100 normal controls. The posterior probability of carrying a breast cancer predisposition gene in the individuals who developed breast cancer was assessed using the Cyrillic genetic risk package, based on the Claus model. 26 women had <50% posterior probability of harbouring a breast cancer predisposition gene and 44 had a ≥ 50% chance of having a breast cancer predisposition gene. In the former group of 26 women, 8 had been taking tamoxifen and 18 placebo. In the group of women with ≥ 50% probability of having a breast cancer gene, 26 had been taking tamoxifen and 18 placebo. The differences between the numbers of women taking tamoxifen who subsequently developed cancer in the two groups divided by <50% or ≥ 50% genetic risk was significant (P = 0.04). These preliminary data suggest that tamoxifen prevention may be more effective in women with a <50% chance of harbouring a breast cancer predisposition gene. A meta analysis of the interaction of genetic status with tamoxifen chemoprevention effectiveness should be conducted to test this hypothesis.