Using the prospective screening Karma cohort, an image-based breast cancer 2-year risk model was developed using mammographic features (density, masses, microcalcifications) identified using artificial intelligence. The model could be extended by including a polygenic risk score with 313 single nucleotide polymorphisms. The area under the receiver operating characteristic curve (AUC) for the image-based model was 0.73 (95% confidence interval [CI]: 0.71, 0.74) and for the genetic extended model was 0.77 (95% CI: 0.75, 0.79). There was a relative 9-fold difference in risk between women at high risk (8% of the population) and those at general risk. High-risk women were more likely to be diagnosed with stage II cancers and with tumours 20 mm or larger and were less likely to have stage I and estrogen receptor-positive tumours.
Identification of Women at High Risk of Breast Cancer Who Need Supplemental Screening. Eriksson et al. Radiology, 2020. DOI: 10.1148/radiol.2020201620