Assessing the Performance of the PREDICT Breast Version 3.0 Prognostic Tool in Patients With Breast Cancer in the...

PREDICT Breast version 3 (v3) is the latest updated prognostication tool, developed using data from approximately 35,000 women diagnosed with breast cancer between 2000 and 2018 in the United Kingdom. Although an earlier version of PREDICT was tested in the United States, the performance of the latest version remains unknown. This study aims to validate PREDICT Breast v3 using newly released SEER outcome data for patients with breast cancer in the United States and to address potential health disparities.

A total of 615,865 female patients diagnosed with primary breast cancer between 2000 and 2018 and followed for at least 10 years were selected from the SEER database. Predicted and observed 10- and 15-year breast cancer-specific survival outcomes were compared for the overall cohort, stratified by estrogen receptor (ER) status and predefined subgroups. Discriminatory accuracy was evaluated using the area under the receiver operating characteristic curve (AUC).

PREDICT Breast v3 demonstrated good calibration and discrimination for long-term breast cancer-specific survival. It provided accurate mortality estimates (within ±10% absolute error) across the US population for 10-year (-10% in ER-positive and 2% in ER-negative breast cancer) and 15-year (4% in ER-positive and 3% in ER-negative breast cancer) all-cause mortality, for both ER statuses. The model also showed good performance for 10- and 15-year all-cause mortality across the US population, with AUC values of 0.769 and 0.794 for ER-positive breast cancer as well as AUC of 0.738 and 0.746 for ER-negative breast cancer, respectively, indicating good discriminatory ability. However, recalibration is needed for specific groups, including non-Hispanic Asian and non-Hispanic Black patients with ER-negative disease.

PREDICT v3 accurately predicts 10- and 15-year breast cancer-specific survival in contemporary US patients with breast cancer. Future efforts should focus addressing disparities observed in predictive tools to promote equitable care.