Abstract
BACKGROUND: Medullary breast carcinoma (MBC) is a rare type of breast cancer. Our study aimed to compare the differences in clinical characteristics and prognosis between MBC and invasive ductal carcinoma (IDC), and to further develop and validate nomograms to predict overall survival (OS) and cancer-specific survival (CSS) in MBC patients. METHODS: A total of 179,613 patients from the Surveillance, Epidemiology and End Results (SEER) database from 2010 to 2015, including 596 MBC patients, were analyzed using the Kaplan-Meier method and propensity score matching (PSM) to compare patients' OS and CSS. Cox proportional hazard regression model was used to determine independent prognostic factors for OS and CSS in MBC patients. Nomograms were constructed based on Cox regression analysis whereas receiver operating characteristic (ROC) curves and calibration curves were used to evaluate the predictive accuracy. RESULTS: There were significant differences in the clinical characteristics between MBC and IDC. According to the logrank test, MBC had better OS and CSS than IDC before and after PSM. Cox multivariate analysis showed that age, race, tumor size, lymph node (LN), and radiation therapy were independent prognostic factors for OS, whereas age, tumor size, American Joint Committee on Cancer (AJCC) stage, laterality, type of surgery, and chemotherapy were independent prognostic factors for CSS. Nomograms of OS and CSS were constructed based on independent prognostic factors. CONCLUSIONS: MBC had better OS and CSS than IDC. Nomograms based on clinicopathological features were sufficiently accurate in predicting the OS and CSS for MBC patients, which can effectively predict the survival risk of MBC patients and guide clinicians to provide more effective treatment measures.