Validating a Model for Predicting Breast Cancer and Non-Breast Cancer Death in Women Aged 55 and Older
Menée à partir de données portant sur 222 044 patientes atteintes d'un cancer du sein, cette étude évalue la performance d'un modèle, basé notamment sur des facteurs clinico-démographiques et des facteurs comportementaux (consommation d'alcool et de tabac, activité physique), pour prédire le risque de cancer du sein à 5 ans et le risque de décès à 10 ans non lié à un cancer du sein chez les femmes âgées de 55 ans ou plus
Background : To support mammography screening decision-making, we developed a competing-risk model to estimate 5-year breast cancer (BC) risk and 10-year non-BC death for women
≥
55 using Nurses’ Health Study (NHS) data and examined model performance in the Black Women’s Health Study (BWHS). Here, we examine model performance in predicting 10-year outcomes in the BWHS, Women’s Health Initiative-Extension Study (WHI-ES) and Multiethnic Cohort (MEC) and compare model performance to existing BC prediction models.
Methods : We used competing-risk regression and Royston and Altman’s methods for validating survival models to calculate our model’s calibration and discrimination (c-index) in BWHS (n = 17,380), WHI-ES (n = 106,894) and MEC (n = 49,668). NHS development cohort (n = 48,102) regression coefficients were applied to the validation cohorts. We compared our model’s performance to BCRAT “Gail” and IBIS models by computing BC risk estimates and c-statistics.
Results : When predicting 10-year BC risk, our model’s c-index was 0.569 in BWHS, 0.572 in WHI-ES, and 0.576 in MEC. BCRAT’s c-statistic was 0.554 in BWHS, 0.564 in WHI-ES, and 0.551 in MEC; IBIS’s c-statistic was 0.547 in BWHS, 0.552, in WHI-ES, and 0.562 in MEC. BCRAT underpredicted BC risk in WHI-ES; IBIS underpredicted BC risk in WHI-ES and in MEC but overpredicted BC risk in BWHS. Our model calibrated well. Our model’s c-index for predicting 10-year non-BC death was 0.760 in WHI-ES and 0.763 in MEC.
Conclusions : Our competing-risk model performs as well as existing BC prediction models in diverse cohorts and predicts non-BC death. We are developing a website to disseminate our model.
Journal of the National Cancer Institute , résumé, 2022