Relationships between clinical and biochemical predictors of chemotherapy-related amenorrhea in premenopausal adjuvant breast cancer patients

Abstract

11011 Background: Premature ovarian failure (POF) is a major side effect of breast cancer adjuvant therapy. Predicting post- chemotherapy POF and reproductive status impacts decisions about fertility, contraception, menopausal issues and breast cancer treatment. We examined clinical and biochemical predictors of chemotherapy related amenorrhea (CRA) in premenopausal adjuvant breast cancer patients. Methods: We conducted a cohort study of 121 breast cancer patients with stages I-III disease, premenopausal at diagnosis, enrolled 1 to 4 years from chemotherapy. All patients received 4 cycles of doxorubicin and cyclophosphomide with or without 4 cycles of paclitaxel or docetaxel, at q2 (dose-dense) or q3 week intervals. The primary endpoint was CRA (amenorrhea > 12 months from chemotherapy start). Pretreatment menstrual status and clinical factors were collected prospectively; subjects were recontacted for additional information and venipuncture. Regression models were used to examine risk factors for CRA and log FSH. Results: Mean age was 43.2 years (range 26.7–57.8). Mean follow up from chemotherapy start to enrollment was 2.1 years (range 1.0–4.1). 68 subjects developed CRA, while 53 continued menstruating. In a model including age, tamoxifen use, dose density, chemotherapy regimen, ovarian suppression during chemotherapy, smoking, body mass index and race, only age (OR 1.57[CI 1.29–1.89], p<0.001), tamoxifen use (OR 7.25 [CI 5.89–34.87], p=0.01), and dose density (OR 3.38 [CI 1.16–16.50], p=0.03) had independent associations with CRA. Serum FSH at enrollment (n=112) was higher in CRA than in non-CRA (11.8 vs 45.8, p< 0.001). However, 39% (25/65) of CRA subjects had FSH < 40 mIU/mL, the biochemical threshold for menopause (mean 26.4, range 10.1–39.9). FSH was not associated with chemotherapy regimen (p=0.51) or schedule (p=0.21). Conclusions: Age, tamoxifen use, dose density and subsequent FSH were independently associated with CRA. Clinically, amenorrhea for more than 12 months did not always represent biochemical menopause. Better methods are needed to characterize ovarian function in patients after adjuvant chemotherapy. No significant financial relationships to disclose.