Prognostic factors and survival prediction in HER2‐positive breast cancer with bone metastases: A retrospective cohort study

BACKGROUND: Bone is the most common metastatic site of breast cancer. The developmental pattern of bone metastasis differs in different molecular subtypes. The prognostic factors of HER2‐positive breast cancer with bone metastases require further investigation. The goal of this retrospective study was to identify the clinical features and prognostic factors for HER2‐positive patients with bone metastases. METHODS: A total of 34,084 HER2‐positive breast cancer cases and 1204 cases of bone metastases from the Surveillance, Epidemiology, and End Results (SEER) database from 2010 to 2015 were analyzed to identify clinical characteristics and prognostic factors. A nomogram was constructed based on the Cox proportional hazards regression model. The C‐index, calibration curve, and receiver operating characteristic (ROC) were utilized for model validation. RESULTS: In the HER2‐positive breast cancer total population (34,084 cases), 6.2% developed metastatic diseases. Bone metastases accounted for 3.5% of the entire cohort and 56.7% of all metastatic cases. Univariate and multivariate Cox regression analyses identified seven prognostic factors for predicting cancer‐specific survival (CSS) for HER2‐positive breast cancer patients with bone metastases, including age, brain metastases, liver metastases, lung metastases, PR status, surgery, and chemotherapy. The C‐index of the nomogram was 0.74 vs. 0.78 (for 3‐year CSS) and 0.77 vs. 0.81 (for 5‐year CSS) in the model and validation cohorts, respectively. The AUCs were 0.74 vs. 0.78 (for 3‐year CSS) and 0.77 vs. 0.81 (for 5‐year CSS) in the model and validation cohorts, respectively. The calibration curves indicated favorable agreement between the actual observations and the predictions. CONCLUSION: Our study provided population‐based clinical features and prognostic factors for HER2‐positive breast cancer patients with bone metastases and we constructed a prognostic nomogram with reliable accuracy.