We examined the effects of a Japanese fermented soybean product, miso, and tamoxifen (TAM), alone and in combination, on N-nitroso-N-methylurea (MNU)-induced rat mammary cancer. Seven-week-old female CD/Crj rats received a single i.v. dose (50 mg/kg body weight) of MNU. After administration of MNU, the rats were divided into 4 groups: regular diet (control), 10% miso diet, regular diet + TAM, and 10% miso diet + TAM. TAM was implanted s.c. in the form of pellets containing 2.5 mg at the same time as MNU was administered. All rats were observed for 18 weeks after MNU administration. Incidence (percentage of rats with tumors) and multiplicity (mean tumors/rat) of mammary tumors were 91% and 4.5 in the control, 77% and 2.4 (P < 0.05) in the 10% miso group, 68% and 1.4 (P <0.01) in the TAM group, and 10% (P < 0.0001 or less) and 0.2 (P < 0.0001) in the 10% miso + TAM group. In the second experiment, the effect of the combination of miso and TAM on established rat mammary tumors was investigated. When the mammary tumors induced by MNU reached 10 to 25 mm, the rats were divided into 3 treatment groups: regular diet, regular diet + TAM, and 10% miso diet + TAM. At 6 weeks after the start of treatment, the mean tumor size in the control and TAM groups was 160% and 141% of the pretreatment value, but a decrease of 85% of the pretreatment value was produced by the combination of miso and TAM, and this was significantly different from both the control and TAM groups (P < 0.01) and P < 0.05, respectively). These results indicate that miso is useful in protecting against mammary cancer and it can be expected to have a potent antitumor effect, especially when used in combination with TAM.