Mammographic breast density has been shown to be associated with up to a 4- to 6-fold increase in risk of breast cancer, whereas tamoxifen therapy increases disease-free survival and reduces mortality. We have therefore investigated whether these effects are related. To determine the effects of tamoxifen on mammographic density, mammograms from 94 women who had received tamoxifen for breast cancer and 188 women (without breast cancer) who had not received tamoxifen were visually classified according to the criteria of Wolfe. Two controls were age-matched to each case. All of the women were postmenopausal (ages, 50-64 years), neither group was taking hormone replacement therapy, and none of the cases had received chemotherapy. There were significant differences in breast density between cases and controls at the initial mammogram (P = 0.0001) but no significant differences at the follow-up mammogram (P = 0.51). A significant change to a more lucent pattern had occurred among the case group (P = 0.0001). The odds ratio for cancer that was associated with the more dense (P2 and DY) patterns with respect to the more lucent (N1 and P1) patterns was 3.6 (95% confidence interval, 2.11-6.18) at the initial mammogram. This was significantly reduced to 1.5 (95% confidence interval, 1.32-1.70) after treatment with tamoxifen (P = 0.019; chi2 = 5.52). The substantial reduction in breast density with tamoxifen provides evidence that tamoxifen has the capacity to favorably alter postmenopausal breast density toward a more lucent pattern, which is associated with reduced risk of breast cancer. Mammographic pattern is, thus, a potential biomarker of breast cancer risk.