A framework for assessing interactions for risk stratification models: the example of ovarian cancer
A l'aide de données issues de 9 études cas-témoins de l'"Ovarian Cancer Association Consortium", cette étude décrit une approche pour évaluer les interactions entre des facteurs de risque et/ou des facteurs protecteurs lors du développement de modèles prédictifs
Generally, risk stratification models for cancer use effect estimates from risk/protective factor analyses that have not assessed potential interactions between these exposures. We have developed a four-criterion framework for assessing interactions which includes statistical, qualitative, biological, and practical approaches. Using ovarian cancer, we present the application of the framework as this is an important step in developing more accurate risk stratification models. Using data from nine case-control studies in the Ovarian Cancer Association Consortium, we conducted a comprehensive analysis of interactions between 15 unequivocal risk/protective factors for ovarian cancer (including 14 non-genetic factors and a 36-variant polygenic score) with age and menopausal status. Pairwise interactions between the risk/protective factors were also assessed. We found that menopausal status modifies the association between endometriosis, first degree family history of ovarian cancer, breastfeeding, and depot-medroxyprogesterone acetate use and disease risk, highlighting the importance of understanding multiplicative interactions when developing risk prediction models.
Journal of the National Cancer Institute , résumé, 2022