Purpose: Ovarian function is important for optimizing endocrine treatment in patients with hormone receptor-positive (HR+) early breast cancer (eBC). The aim of the study was to determine whether patients' pretreatment levels of anti-Mullerian hormone (AMH) were associated with menses status after chemotherapy and to build a predictive nomogram model for amenorrhea in women with HR+ eBC.
Methods: Between August 2013 and December 2014, 120 premenopausal patients with HR+ eBC were included retrospectively. The associations among age, prechemotherapy levels of AMH, follicle-stimulating hormone (FSH),and estradiol (E2) and the 2-year postchemotherapy menses status were analyzed. We determined the cutoff values of hormone levels by using the biostatistical tool (Cutoff Finder). A novel nomogram was established to predict the 2-year amenorrhea status based on the logistic analysis. Concordance index (C-index) was used to validate the capacity.
Results: One hundred nine women (90.8%) experienced amenorrhea after chemotherapy. AMH < 0.965 ng/ml predicted amenorrhea at 2 years (AUC 0.84, sensitivity 74% and specificity 81.8%), independent of age. The predictive nomogram based on age and pretreatment AMH and FSH levels was developed to predict the probability of 2-year postchemotherapy amenorrhea with a C-index of 0.88 (95% CI 0.84-0.91).
Conclusions: In premenopausal patients with HR+ eBC, prechemotherapy AMH concentration was associated with the patient's 2-year amenorrhea status, independent of age. The nomogram model based on age and pretreatment AMH and FSH levels accurately predicted the 2-year amenorrhea status.
Keywords: Amenorrhea; Anti-Mullerian hormone; Breast cancer; Chemotherapy; Nomogram.