Predictive factors of adjuvant therapy-related amenorrhea for patients with breast cancer.

Abstract

217 Background: Reproductive age breast cancer patients often experience therapy-related amenorrhea so preserving fertility after treatment is a major concern. Anti-mullerian hormone (AMH) values have been reported useful in predicting menopause for infertility treatment patients, but published information is limited regarding breast cancer patients. If amenorrhea can be predicted before treatment, breast cancer patients hoping to have children in the future can consider this information in choosing treatment. Our aim was to predict patient fertility using AMH values. Methods: Forty breast cancer patients, 25-45 years of age (median, 35.5), who received adjuvant chemotherapy (CT) or endocrine therapy (ET), had AMH values retrospectively assayed from frozen serum before and after treatment. AMH values were then analyzed in relation to clinicopathological factors using logistic analysis. Of 19 adjuvant CT patients whose AMH values could be assayed, Cox’s proportional hazard model indicated a correlation in terms of amenorrhea and time to menses resumption between clinicolpathological factors and AMH values. Results: The mean pretreatment AMH value for all 40 patients was 21.4pM (range, 3-78) and a correlation was demonstrated with patient age, but not breast cancer intrinsic subtype or stage. AMH values decreased to the lower limit after treatment for all CT patients regardless of age, but there was a difference in degree depending on patient age for ET patients. Of the 19 CT patients, the mean AMH value was 25.6pM (3-78). These patients all underwent a regimen of anthracycline with 11 (58%) also receiving a regimen of taxiane. After starting CT, 17 patients (89%) experienced amenorrhea, but 11 (59%) resumed menstruation within one year of finishing CT and 15 (78%) resumed within two years. There were no significant differences among CT patients in age, pretreatment AMH value, intrinsic subtype or stage, but patients receiving taxiane demonstrated a noteworthy trend. Conclusions: Although we were unable to predict menopause directly using AMH values given the small number of patients in this study, we could estimate the risk of menopause as being either low or high for CT patients based on age and type of regimen.