Background With the prolonged overall survival of breast cancer (BC) patients, the incidence of breast cancer brain metastasis (BCBM) is gradually increasing. However, the prognosis of BCBM patients remains significantly poor.
Objective To analyze prognostic factors of patients with BCBM, establish and evaluate a clinical nomogram for predicting prognosis of BCBM patients.
Methods Clinical data about patients diagnosed with BCBM at our center from January 2003 to June 2021 were retrospectively collected. Prognostic factors were determined using Cox regression analysis, and a prognostic nomogram model was developed to predict the survival of BCBM patients at 6 months, 1 year, and 2 years. The predictive performance and calibration of the nomogram were evaluated using ROC curves and calibration plots, with internal validation performed using bootstrap methods.
Results A total of 700 patients were enrolled. The mean age at diagnosis of BC was 43.77 years, and the median overall survival of patients was 11 months (95% CI: 9.8-12.2 months). Multivariable Cox regression analysis revealed that poor prognosis in BCBM patients was independently associated with older age (>48 years) at BC diagnosis (HR=1.30, 95% CI: 1.09-1.56, P=0.003), Karnofsky Performance Status (KPS) <80 at BCBM diagnosis (HR=1.34, 95% CI: 1.13-1.60, P=0.001), triple-negative breast cancer (HR-positive/HER2-negative: HR=0.77, 95% CI: 0.60-0.98; HER2-positive: HR=0.62, 95% CI: 0.50-0.77; P<0.001), non-brain site of first metastasis (HR=1.83, 95% CI: 1.44-2.32, P<0.001), number of brain metastases≥3 (HR=1.44, 95% CI: 1.20-1.72, P<0.001), leptomeningeal metastasis (HR=1.53, 95% CI: 1.12-2.08, P=0.008), not receiving radiotherapy after BCBM diagnosis (HR=1.31, 95% CI: 1.10-1.57, P=0.003), liver metastasis (HR=1.28, 95% CI: 1.07-1.52, P=0.006) and lung metastasis (HR=1.19, 95% CI: 1.00-1.42, P=0.049). A prognostic nomogram of BCBM patients was constructed, with area under the curve (AUC) values of 0.723, 0.720, and 0.798 at 6 months, 12 months, and 24 months, respectively. Internal validation demonstrated an average AUC >0.7 for each time period, the average AUCs and 95%CIs for predicting patients’ survival at 6 months, 1 year and 2 years were 0.725 (0.723-0.727), 0.723 (0.721-0.725) and 0.759 (0.757-0.761), respectively. Calibration and decision curves indicated great accuracy and clinical utility of the model.
Conclusion We have developed a prognostic nomogram with good performance for survival in BCBM patients. Age at BC diagnosis, KPS at BCBM diagnosis, molecular subtype, first metastatic site, number of brain metastases, presence of leptomeningeal metastasis, receipt of radiotherapy, presence of liver metastasis and lung metastasis are independent prognostic factors for breast cancer patients with brain metastasis.
DownLoad: