Retrospective Cohort Study on the Limitations of Direct-to-Consumer Genetic Screening in Hereditary Breast and Ovarian Cancer.

Purpose: Among cancer predisposition genes, most direct-to-consumer (DTC) genetic tests evaluate three Ashkenazi Jewish (AJ) founder mutations in , which represent a small proportion of pathogenic or likely pathogenic variants (PLPV) in cancer predisposing genes. In this study, we investigate PLPV in and other cancer predisposition genes that are missed by testing only AJ founder mutations.

Methods: Individuals were referred to genetic testing for personal diagnoses of breast and/or ovarian cancer (clinical cohort) or were self-referred (nonindication-based cohort). There were 348,692 participants in the clinical cohort and 7,636 participants in the nonindication-based cohort. Both cohorts were analyzed for AJ founder mutations. Full sequence analysis was done for PLPV in , , , , , , , , , (truncating variants), , , , , , , and 22 other genes.

Results: AJ founder mutations accounted for 10.8% and 29.7% of PLPV in the clinical and nonindication-based cohorts, respectively. AJ founder mutations accounted for 89.9% of PLPV in those of full AJ descent, but only 69.6% of those of partial AJ descent. In total, 0.5% of all individuals had a AJ founder variant, while 7.7% had PLPV in a high-risk breast/ovarian cancer gene. For non-AJ individuals, limiting evaluation to the AJ founder mutations missed >90% of mutations in actionable cancer risk genes. Secondary analysis revealed a false-positive rate of 69% for PLPV outside of non-AJ founder mutations.

Conclusion: DTC genetic testing misses >90% of PLPV in individuals of non-AJ ancestry and about 10% of PLPV among AJ individuals. There is a high false-positivity rate for non-AJ PLPV with DTC genetic testing.