Objective: Determine whether differences in serum hormone concentrations explain the lack of NAF yield among women participating in a study of NAF hormone concentrations in breast cancer cases and screening controls.

Methods: 80 NAF yielders (Y) and 80 non-yielders (NY) were randomly selected from women who had completed participation in an ongoing case-control study, equally divided by menopausal status and case-control status (only the contralateral breast was aspirated in cases). Average age of premenopausal women was 45.95 years and of postmenopausal women was 57.54 years. The percentage of African American and Caucasian subjects was similar among those who yielded NAF and those who did not. Serum was collected on the same day that nipple aspiration was attempted and the serum was assayed for estradiol, progesterone, and FSH by standard immunoassays. Hormonal concentrations were compared among 80 subjects who produced >2 microliters of NAF and 80 who did not. Wilcoxon tests were used to compare rank distributions of the values for yielders and non-yielders within each menopausal and case/control category.

Results: Hormonal data were analyzed in quintiles with the following comparisons: premenopausal cases Y vs NY; postmenopausal cases Y vs NY; premenopausal controls N vs NY; postmenopausal controls N vs NY. Among premenopausal cases estradiol was significantly greater at the P = 0.0067 level among those who did not produce NAF. This remained significant after correction for multiple comparisons. The difference remained significant after combining cases and controls at the P = 0.021 level. The other comparisons were not significant. Analyses of progesterone and FSH yielded no significant differences.

Discussion: The inhibition of breast fluid accumulation by estradiol is consistent with the effect of ovarian steroids on milk secretion observed after parturition. The observation that this effect was greater in cases is interesting and will be investigated further to ascertain whether other factors associated with breast cancer may play a role in the observed inhibition.

Citation Information: Cancer Prev Res 2010;3(12 Suppl):B33.