Virtual versus reality: Assessing the accuracy of a digital cancer risk assessment tool in a gynecologic oncology clinic.
Academic Article
Overview
abstract
OBJECTIVE: Integrating comprehensive cancer history assessment and risk stratification into clinical practice is challenging and underutilized. Digital tools leveraging validated risk models have the potential to assist clinicians with cancer risk assessment. This study evaluates a digital cancer risk assessment tool in a gynecologic oncology clinic. METHODS: Patients presenting for a gynecologic oncology appointment completed a digital personal and family history-based cancer risk assessment prior to their appointment. The tool assessed eligibility for genetic testing using National Comprehensive Cancer Network (NCCN) criteria and estimated lifetime breast cancer risk using the Tyrer-Cuzick (TC) model. During the visit, clinicians reviewed the patient's personal/family history and risk assessment to assess tool accuracy. RESULTS: Two-hundred gynecologic oncology patients completed the digital cancer risk assessment tool between November 2022 and February 2024. Median age was 45 years (interquartile range 21). race and ethnicity Non-Hispanic White (137; 68.5 %), Asian (22; 11.0 %), Hispanic (15; 7.5 %), Black or African American (7; 3.5 %), Native Hawaiian or Other Pacific Islander (1; 0.5 %), or Other (8; 4.0 %). The digital cancer risk assessment tool accurately identified patients meeting NCCN criteria for genetic testing with sensitivity of 94.0 % and specificity of 100 %. The tool accurately identified patients with an elevated TC breast cancer risk score with a sensitivity of 91.1 % and specificity of 97.0 %. CONCLUSIONS: A digital cancer risk assessment tool demonstrated high sensitivity and specificity in identifying patients eligible for genetic testing and enhanced breast screening. Digital tools may represent a reliable and scalable mechanism to enhance cancer risk recognition in clinical settings.
