Sex hormone concentrations and the risk of breast cancer recurrence in postmenopausal women without hot flashes

Abstract

Background: We examined if the reduced risk of breast cancer events seen among women without baseline hot flash symptoms in the Women's Healthy Eating and Living (WHEL) dietary intervention trial was related to changes in sex hormone concentrations.

Methods: Baseline and year one concentrations of total and bioavailable estradiol, and testosterone and sex hormone-binding globulin (SHBG) were compared by intervention arm among 447 postmenopausal women without hot flashes. Cox proportional hazard models tested interaction terms between study arm and baseline hormone concentrations adjusted for study site, antiestrogen use, positive nodes, tumor size, oophorectomy status, and hormone replacement therapy use.

Results: Sex hormone concentrations did not differ by study arm at baseline nor at year one. Twenty-two (9.8%) events occurred in the intervention arm versus 42 (18.9%) in the comparison arm (P = 0.009). Baseline bioavailable testosterone was significantly, positively associated with additional events (HR 1.69, 95% CI: 1.00-2.84; P = 0.049). There were significant interactions between the intervention and total (P = 0.015), and bioavailable (P = 0.050) testosterone: the intervention was more protective among participants with higher baseline total (HR 0.3, 95% CI: 0.2-0.7) or bioavailable (HR 0.4, 95% CI: 0.2-0.7) testosterone than for participants with lower baseline total (HR 0.8, 95% CI: 0.4-1.5) or bioavailable (HR 0.8, 95% CI: 0.4-1.5) testosterone. No significant effects were seen for estradiol or SHBG.

Conclusions: The WHEL dietary intervention may have modified other risk factors of recurrence correlated with testosterone.

Impact: Sex hormones should be considered as part of a larger biological system related to the risk of breast cancer recurrence.

©2011 AACR.

Figures

Figure 1

Risk reduction for intervention arm versus comparison arm by baseline testosterone concentrations (ng/dL): postmenopausal women who did not report hot flash symptoms at baseline. Hazard ratios with 95% confidence intervals for intervention arm versus comparison arm computed using models 4 and 5 from Table 2, inserting values of baseline hormone concentrations representing the 25th, 50th, and 75th percentiles as examples. Models adjusted for site, anti-estrogen use, number of positive nodes, tumor size, oophorectomy status, and previous hormone replacement therapy use. Sample size: n=22 (9.8%) events out of 225 participants in the intervention arm versus n=42 (18.9%) events out of 222 participants in the comparison arm; subset excludes women who recurred within the first year of the WHEL study.