Background

Genetic screening of unaffected individuals for hereditary breast and ovarian cancer (HBOC) risk is a growing opportunity for personalized preventive medicine. The FDA recently authorized a direct-to-consumer (DTC) test to report on 3 BRCA1/2 variants, commonly found in individuals of Ashkenazi Jewish (AJ) heritage, out of more than 1000 known. Here we define the probability that DTC genetic screening for the 3 BRCA1/2 AJ founder variants would falsely reassure individuals of AJ and non-AJ ancestry of low risk for hereditary cancer syndromes (HCS). We also assess the frequency of false positive results reported by third parties from raw genotypes, which are routinely provided to clients who submit them for cancer risk analysis.

Methods

We analyzed de-identified data on three cohorts: 1) An indication-based cohort of 119,328 patients referred by healthcare providers for HBOC genetic testing due to personal or family history, 2) a screening cohort of 5,170 patients without personal or family history who had BRCA1/2 testing as a health screen, and 3) a confirmation cohort of 102 patients referred for clinical confirmatory testing instigated by positive DTC results from third party analysis of raw data.

Results

In the indication-based cohort 12,846 patients had a mutation in any HCS-associated gene: 4,733 (37%) were in BRCA1/2, of which 12% were one of the 3 AJ founder mutations. Ethnicity impacted AJ founder mutation frequency: 81% of AJ patients with any BRCA1/2 mutation had one of the 3 founder mutations, but only 6% of non-AJ patients. Overall clinical false-negative rate for the 3 AJ founder mutations in BRCA1/2 carriers was 88%; rates were 19% and 94% among AJ and non-AJ individuals, respectively. In the screening cohort, where 2.6% were AJ, 40 patients had a P/LP mutation in BRCA1/2; 12.5% were an AJ founder mutation. The clinical false-negative rate of 88.5% for any BRCA1/2 mutations in the screening cohort is similar to the indication-based cohort. Finally, in the confirmation cohort, among patients told of a positive DTC screening result, our analyses indicated 50% (52/102) were analytic false positives.

Conclusions

This study highlights the limitations of screening restricted to a few of the many genetic variants associated with HBOC risk. We found a greater rate of non-founder BRCA1/2 mutations in AJ patients than previously reported, and found a concerning rate of clinical false negatives with screening limited to the AJ founder variants. Although increasing access to genetic information is vital, the limitations of current DTC genetic screening, despite warnings from the FDA, may not be well understood by consumers and should be used with caution. Patients screened for HBOC on a platform limited to the AJ founder variants and supplemented by analysis of raw genotyping data, should receive confirmatory testing, regardless of having either a positive or negative result, a recommendation which the DTC companies and third party analysts of raw data ostensibly support. These results suggest that all HCS screening should include the support of a qualified clinician to assess the potential limitations and implement appropriate clinical management recommendations for patients and their family members.

Citation Format: Esplin ED, Haverfield E, Yang S, Herrera B, Anderson M, Nussbaum RL. Limitations of direct-to-consumer genetic screening for HBOC: False negatives, false positives and everything in between [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P4-03-06.