Validation of a Digital Tool That Uses National Testing Guidelines to Identify Individuals at Risk for Hereditary Cancer.

Background: NCCN publishes genetic testing criteria based on personal and family cancer history (PFHx). Digital risk stratification tools may aid clinicians in the challenging task of systematically collecting this history and accurately identifying individuals who meet the criteria. A HIPAA-compliant digital tool (The Ambry CARE Program) integrates NCCN Guidelines and identifies patients who meet the NCCN criteria for hereditary breast, ovarian, pancreatic, and prostate cancer (HBOP), as well as Lynch syndrome and familial adenomatous polyposis (both related to hereditary colorectal cancer [CRC]). The purpose of this study was to validate the analytic accuracy of the tool's interpretation of the NCCN Guidelines compared with a certified genetic counselor (CGC).

Methods: This study, conducted by a diagnostic laboratory, included 2 phases: (1) development and internal verification of the tool using 1,300 theoretical clinical scenarios (913 HBOP and 394 CRC scenarios), with testing eligibility determined by internal CGC consensus and the tool; and (2) external analytical validation comparing CARE's interpretation of 400 deidentified real-world cases against CGC interpretation. Of the 400 cases, 200 met and 200 did not meet the criteria. Of the cases that met the criteria, based on CARE's interpretation, 150 met the NCCN Guidelines for Genetic/Familial High-Risk Assessment: Breast, Ovarian, and Pancreatic (Version 2.2022), and 50 met the NCCN Guidelines for Genetic/Familial High-Risk Assessment: Colorectal (Version 1.2021).

Results: There were no discrepancies in the internal verification when comparing the final consensus outcomes with the CARE outputs. In external validation, CARE accurately assessed 398 (99.5%) cases. In 17 cases, CARE made correct risk assessments, whereas CGCs did not.

Conclusions: CARE accurately identifies individuals who meet NCCN testing criteria to aid in risk stratification. Digital tools such as this may be helpful in clinical practice for collecting PFHx and identifying candidates for genetic testing of hereditary cancers.