Choosing Breast Cancer Risk Models: Importance of Independent Validation
Menée à partir de données portant sur 35 921 femmes ayant subi une mammographie de dépistage (âge : de 40 à 84 ans), cette étude compare la performance de cinq modèles pour prédire le risque de cancer du sein
Several widely used risk models estimate the chance that a woman will develop breast cancer over a defined time interval such as 5 years. Before usingsuch a model, one should consider the nature of the target population and whether the modelhasbeen validated for that population. In this issueof Journal of the National Cancer Institute, McCarthy and colleagues (1)present valuable data comparing the performance of fiverisk models in 35,921 predominantly white women who came to the Newton-Wellesley Hospital in Massachusetts for mammographic screens between 2007-2009. This study adds importantly to two other recent large validation studies. The U.K. Generation Study cohort (2)of64,874womenwas recruited between 2003-2012 from the general U.K. population. The Breast Cancer Prospective Family Study Cohort(3)(ProF-SC) included 15,732 womenrecruited from Australia, Canada and the U.S. between 1992-2011from high-risk clinics oras relatives of women inbreast cancer registries. Of the women in ProF-SC, 82% had at least one affected first-degree relative, and 6.83% carried a mutation in BRCA1 or BRCA2(4). These two general population cohorts and one high-risk cohort enable assessment of risk model performance.
Journal of the National Cancer Institute , éditorial en libre accès, 2018