Reproductive history and prognosis in patients with operable breast cancer.

BACKGROUND Late menarche, early menopause, high parity, and early first birth decrease the risk of development of breast cancer. The influence of these factors on the survival of breast cancer patients has not been explained. METHODS A group of 1885 patients with operable breast cancer was studied retrospectively. A univariate analysis was used to calculate 10-year overall survival (OS) and disease free survival (DFS) in relation to age, menopausal status, age at menarche and menopause, and number of pregnancies and deliveries. A multivariate analysis (Cox model) was performed in which classic prognostic factors (tumor size and grade, lymph node involvement) were included in addition to reproductive factors. RESULTS Univariate analysis demonstrated better prognosis in patients who had never been pregnant compared with those who had (OS, 62% vs. 54%, respectively; P = 0.01; DFS, 53% vs. 44%, respectively; P = 0.005) and in nulliparous compared with parous patients (OS, 62% vs. 53%, respectively; P = 0.006; DFS, 52% vs. 44%, respectively; P = 0.004). Survival rates decreased with the number of pregnancies and deliveries. Patients with late menarche had worse survival then those whose first menstruation occurred before the age of 16 years (DFS, 47% vs. 41%, respectively; P = 0.04). By multivariate analysis, parity remained an independent prognostic indicator in addition to classic highly significant prognostic factors (nodal involvement, tumor grade and size). CONCLUSIONS Results suggest that reproductive factors known to decrease the risk of breast cancer development have an adverse effect on prognosis.