Urine estrogens, frequency of ovulation, and breast cancer risk: case-control study in premenopausal women

Urine specimens from 94 premenopausal women with breast cancer and from 70 control women have been compared with respect to concentration of the three major estrogen fractions and to frequency of ovulation as assessed by urine pregnanediol. The probability of anovulation (0.14 in the breast cancer patients and 0.09 among the controls) was not significantly higher among the women with breast cancer (P approximately 0.30). However, there was a positive association between urine estrogen concentration and breast cancer risk. The association was statistically significant (P less than 0.05) for each of the three estrogens measured and in both the follicular and the luteal phases of the menstrual cycle; the relative risk increased from 1 in the referent category (less than 5 micrograms estrogen/g creatinine) to about 3 in the highest category (greater than or equal to 15 micrograms estrogen/g creatinine). The association between urine estrogens and breast cancer risk was consistently stronger when the comparison was restricted to specimens collected in menstrual cycles during which ovulation occurred.