Epidemiologic data and animal experiments suggest that dietary fat may influence risk of breast cancer. To determine whether intervention with a low-fat, high-carbohydrate diet would reduce breast cancer incidence in women at increased risk of the disease, we carried out a randomized controlled trial in Canada. We recruited 4,690 women with extensive mammographic density and randomized them to an intervention group or a comparison group. The intervention group received intensive dietary counseling to reduce fat intake to a target of 15% of calories and increase carbohydrate to 65% of calories. Dietary intakes were assessed throughout using food records. Subjects were followed for at least 7 years and for an average of 10 years. The main outcome was invasive breast cancer. Percentage of calories from fat in the intervention group decreased from 30% at baseline to 20% after randomization and remained 9% to 10% lower than the comparison group throughout. There were 118 invasive breast cancers in the intervention group and 102 in the comparison group [adjusted hazard ratio = 1.19 (95% CI: 0.91-1.55)]. Analysis of food records showed that fat intake at baseline and after randomization was not associated with total breast cancer incidence. Greater weight and lower carbohydrate intake at baseline and after randomization were associated with an increased risk of estrogen receptor (ER)-positive breast cancer. Our findings suggest that a sustained reduction in dietary fat intake did not reduce risk of breast cancer in women with extensive mammographic density. Weight and carbohydrate intakes were associated with risk of ER-positive breast cancer.