Calculating future 10-year breast cancer risks in risk-adapted surveillance: A method comparison and application in clinical practice.

The German Consortium for Hereditary Breast and Ovarian Cancer (GC-HBOC) has successfully implemented risk-adapted breast cancer surveillance for women at high breast cancer risk in Germany. Women with a family history of breast and ovarian cancer, but without pathogenic germline variants in recognized breast cancer risk genes, are recommended annual breast imaging if their predicted 10-year breast cancer risk is 5% or higher, using the BOADICEA BC risk model, as outlined in the current GC-HBOC guideline. However, women who initially do not meet this risk threshold may do so later, even if there is no new cancer in their family. To determine when this threshold is crossed, one could annually repeat BOADICEA calculations using an aging pedigree: the 'prediction by aging pedigree' (AP) approach. Alternatively, we propose a simplified and more practical 'conditional probability' (CP) approach which calculates future risks based on the initial BOADICEA assessment. Using data from 6,661 women registered with GC-HBOC, both methods were compared. Initially, 74% of women aged 30 to 48 years had a 10-year breast cancer risk below 5%, but 53% exceeded this threshold at an older age based on the AP approach. Among the women with an initial risk below the threshold, the CP approach revealed that 99% of women exceeded the 5% threshold at the same or an earlier age compared with the AP approach (88% of cases were within the same year or one year earlier). The CP approach has been implemented as a user-friendly web application.