BMC Women's Health (Jul 2023)

Development and validation of competitive risk model for older women with metaplastic breast cancer

  • Jie Tang,
  • Dianlong Zhang,
  • Xiudan Pan

DOI
https://doi.org/10.1186/s12905-023-02513-x
Journal volume & issue
Vol. 23, no. 1
pp. 1 – 11

Abstract

Read online

Abstract Background Metaplastic breast cancer (MpBC) is a rare histological subtype of breast cancer. This study aims to establish a competitive risk model for older women with MpBC to predict patients’ survival accurately. Methods Data on patients diagnosed with MpBC from 2010 to 2019 are from the Surveillance, Epidemiology and End Results (SEER) program in the United States. All patients were randomly assigned to the training set and validation set. The proportional sub-distribution risk model was used in the training set to analyze the risk factors affecting patient death. Based on the risk factors for cancer-specific mortality (CSM) in patients, we constructed a competitive risk model to predict patients’ 1-, 3-, and 5-year cancer-specific survival. Then we used the concordance index (C-index), the calibration curve and the area under the receiver operating characteristic curve (AUC) to validate the discrimination and accuracy of the model. Results One thousand, four hundred twelve older women with MpBC were included in this study. Age, T stage, N stage, M stage, tumor size, surgery and radiotherapy were risk factors for CSM. We established a competitive risk model to predict 1-, 3-, and 5-year cancer-specific survival in older women with MpBC. The C-index of the model was 0.792 in the training set and 0.744 in the validation set. The calibration curves in the training and validation sets showed that the model’s predicted values were almost consistent with the actual observed values. The AUC results show that the prediction model has good accuracy. Conclusion We developed a competitive risk model based on these risk factors to predict cancer-specific survival in older women with MpBC. The validation results of the model show that it is a very effective and reliable prediction tool. This predictive tool allows doctors and patients to make individualized clinical decisions.

Keywords