Risk factor analysis for human epidermal growth factor receptor 2-positive breast cancer patients with brain metastasis: A risk prediction model based on the Surveillance, Epidemiology, and End Results database

Introduction: Human epidermal growth factor receptor 2 (HER2)-positive breast cancer is characterized by increased cell proliferation and a high potential for metastasis, resulting in a poorer prognosis–especially when brain metastases occur. At present, no definitive clinical factors or predictive models exist to guide prognosis in this patient group. Objective: This study aims to identify prognostic factors affecting survival in HER-2 breast cancer patients with brain metastases and to construct a model to assess prognosis based on clinical characteristics. Methods: Data from female patients diagnosed with breast cancer brain metastases between 2011 and 2020 were extracted from the Surveillance, Epidemiology, and End Results database. Seventeen clinical factors were evaluated for their association with brain metastasis in HER2-positive breast cancer. Univariate and multivariate Cox proportional hazard regression analyses were performed, and Kaplan-Meier survival curves were plotted based on independent risk factors. A nomogram model was then developed to predict survival rates at different time intervals. Model performance was evaluated using calibration curves and the concordance index. Receiver operating characteristic analysis and decision curve analysis (DCA) were also conducted to evaluate predictive accuracy and clinical utility. Results: Cox regression analyses identified nine clinical characteristics significantly associated with overall survival (OS) and seven factors influencing breast cancer-specific survival: molecular subtype, age, node stage, liver and lung metastases, and estrogen receptor and progesterone receptor status. The models achieved concordance indices of 0.680 (OS) and 0.663 (breast cancer-specific survival). The nomogram demonstrated strong predictive power, with area under the curve values of 0.767 and 0.700, respectively. DCA showed that the model curve significantly outperformed the two extreme curves. Conclusion: This study identified key prognostic factors and developed a nomogram-based model with strong predictive value for HER2-positive breast cancer patients with brain metastases.