The adoption of multi-gene panel testing (MGPT) has been a game changer in the hereditary breast cancer arena. As evidence surrounding the growing number of breast cancer predisposition genes has accumulated, ATM, CHEK2, and PALB2 have stood out and become undisputed susceptibility genes conferring a moderate risk for breast cancer. Studies showing consistent and convincing relative risks have led the National Comprehensive Cancer Network (NCCN) to develop surveillance guidelines for individuals with pathogenic variants in these genes. These efforts have allowed us to begin answering the question, “How do I manage these patients once they are identified?” However, many questions remain unanswered, including, “How do I identify these individuals in the first place?” In our recent study, we set out to assess the clinical sensitivity of existing NCCN BRCA1/2 testing criteria when applied to ATM, CHEK2, and PALB2.
Using our sample of 89,008 female breast cancer patients who underwent MGPT from June 2012 to December 2016, we found that the overall clinical sensitivity of the breast-specific criteria was 92.4% for ATM (785 of 850 positive individuals met at least one criteria), 91.3% (1590) for CHEK2, 93.1% (656) for PALB2, and 95.7% (2951) for BRCA1/2. Criteria without an age limit for breast cancer diagnosis performed better for ATM, CHEK2, and PALB2 than for BRCA1/2, which may reflect the difference in age-related penetrance. In contrast, criteria based on features known to be associated with BRCA1 (such as triple negative breast cancer) underperformed for ATM, CHEK2, and PALB2.
Overall, the NCCN BRCA1/2 testing criteria demonstrated high clinical sensitivity when applied to ATM, CHEK2, and PALB2. This study is a necessary first step in exploring the appropriateness of applying BRCA1/2 testing guidelines to moderate penetrance breast cancer genes. Our hope is that by sharing these findings, we may help systematically identify these at-risk individuals which then allows clinicians to give the patients the care they need.