Soy consumption may be related to lower breast cancer risk as assessed by breast density. The aims of this 2-y trial were to examine the effects of soy foods and lifetime soy intake on mammographic density. After 220 premenopausal women were randomly assigned to the intervention or control group, the former group consumed 2 daily servings of soy foods equivalent to 50 mg of isoflavones and the latter consumed their regular diet. The respective dropout rates were 15.6 and 12.6%; adherence to the study regimen was high. We assessed lifetime soy intake with a questionnaire and measured breast density in screening mammograms obtained at baseline and at the end of the trial for 98 intervention and 103 control women using a computer-assisted method. None of the mammographic outcomes differed significantly by experimental group. The total area of the breast increased and the size of the dense areas decreased significantly over time in both groups. After 2 y, the mean percentage density had decreased by 2.8 and 4.1% in intervention and control women, respectively. Women who reported eating more soy during their lives had higher percentage densities than women whose diet included little soy; this difference was significant only in Caucasians. Lower soy intake during early life and higher soy intake during adulthood predicted a greater reduction in the percentage density during the study period. After 2 y of intervention, we observed no significant differences in mammographic densities by intervention status, but it appears that soy consumption throughout life may have some effect on breast density.