Abstract
BACKGROUND: The association between body mass index (BMI) at the time of breast cancer diagnosis and the prognosis of breast cancer patients remains controversial. Furthermore, the association between BMI and prognosis with respect to different breast cancer subtypes is not clearly defined. METHODS: We analyzed data from 41,021 invasive breast cancer patients between January 1988 and February 2008 from the Korean Breast Cancer Registry (KBCR) database. Overall survival (OS) and breast cancer-specific survival (BCSS) were analyzed using the Kaplan-Meier method and Cox's proportional hazard regression model among all patients and specific breast cancer subtypes with respect to BMI categories. RESULTS: A U-shaped association between BMI and mortality was observed in the total cohort. Underweight and obese individuals exhibited worse OS (hazard ratio, 1.23 [95 % confidence interval {CI}, 1.05 to 1.44] and 1.29 [1.13 to 1.48], respectively) and BCSS (1.26 [1.03 to 1.54] and 1.21 [1.02 to 1.43], respectively) than normal-weight individuals. In the estrogen receptor (ER) and/or progesterone receptor (PR)+/human epidermal growth factor receptor 2 (HER2) - subgroup, obese individuals exhibited worse OS (1.48 [1.18 to 1.85]) and BCSS (1.31 [1.13 to 1.52]) than normal-weight individuals. Conversely, in the ER and PR-/HER2+ subgroup, underweight individuals exhibited worse OS (1.68 [1.12 to 2.47]) and BCSS (1.79 [1.11 to 2.90]) than normal-weight individuals. CONCLUSIONS: We observed a U-shaped relationship between BMI at diagnosis and poor OS and BCSS among all breast cancer patients. However, obesity in the ER and/or PR+/HER2- subgroup and underweight in the ER and PR-/HER2+ subgroup were poor prognostic factors. Therefore, BMI at diagnosis and breast cancer subtype should be considered simultaneously in various treatment decision processes and surveillance schedules.