BRCA1 and BRCA2 mutation carrier predictions using the BRCAPRO model in clinic-based minority families enrolled in the Breast Cancer Family Registry (B-CFR)

Abstract

21037 Background: BRCA mutation prediction models, such as BRCAPRO, are used in cancer risk clinics, but they were developed based on mutation rates and penetrance observed in individuals of Ashkenazi Jewish and European ancestry. Furthermore, in clinical counseling it is not always clear whether to test additional family members for BRCA mutations after the proband tests negative. The aim of this study is to evaluate the performance of the BRCAPRO model among clinic-based minority families and to evaluate the clinical utility of testing additional family members in high risk families. Methods: A total of 314 families with at least one member having been tested for BRCA mutations were enrolled through the B-CFR. Families of Ashkenazi Jewish ancestry were excluded. Using the BRCAPRO model with default penetrance and allele frequency for non-Ashkenazi Jewish populations, the predicted likelihood of carrying either a BRCA1 or BRCA2 mutations was generated. Sensitivity, specificity, and area under the receiver operating characteristic curves (AUC) were calculated. Results: There were 55 African American, 181 Hispanic, 42 Asian American and 36 other minority families. The AUC was 0.771 (95% confidence interval, CI: 0.720–0.816) for all minorities combined. At a predicted probability of 10%, the sensitivity for identifying mutation carriers was 65% and the specificity was 76%. The performance was marginally significantly better for Hispanic families than the other minority groups (p=0.07). In 228 families (73%), multiple individuals were tested for BRCA mutations. After taking into account the mutation status of family members, the performance of the BRCAPRO model was improved: the AUC increased to 0.862 (95% CI: 0.820–0.899) and the specificity at the 10% threshold increased to 83%, while the sensitivity remained the same. Conclusions: The data support the use of BRCAPRO in pretest prediction of BRCA1/2 mutations for minority families in high risk clinic settings. Mutation status of family members provides additional predictive value, which may help counselors decide whether to offer other family members the test when one member has already tested negative, given a positive family history of breast and ovarian cancer. No significant financial relationships to disclose.