Predicting advanced neoplasia at colonoscopy in a diverse population with the National Cancer Institute colorectal cancer risk‐assessment tool
Open Access
Abstract
BACKGROUND: Tailoring screening to colorectal cancer (CRC) risk could improve screening effectiveness. Most CRCs arise from advanced neoplasia (AN) that dwells for years. To date, no available colorectal neoplasia risk score has been validated externally in a diverse population. The authors explored whether the National Cancer Institute (NCI) CRC risk‐assessment tool, which was developed to predict future CRC risk, could predict current AN prevalence in a diverse population, thereby allowing its use in risk stratification for screening.METHODS: This was a prospective examination of the relation between predicted 10‐year CRC risk and the prevalence of AN, defined as advanced or multiple (≥3 adenomatous, ≥5 serrated) adenomatous or sessile serrated polyps, in individuals undergoing screening colonoscopy.RESULTS: Among 509 screenees (50% women; median age, 58 years; 61% white, 5% black, 10% Hispanic, and 24% Asian), 58 (11%) had AN. The prevalence of AN increased progressively from 6% in the lowest risk‐score quintile to 17% in the highest risk‐score quintile (P = .002). Risk‐score distributions in individuals with versus without AN differed significantly (median, 1.38 [0.90‐1.87] vs 1.02 [0.62‐1.57], respectively; P = .003), with substantial overlap. The discriminatory accuracy of the tool was modest, with areas under the curve of 0.61 (95% confidence interval [CI], 0.54‐0.69) overall, 0.59 (95% CI, 0.49‐0.70) for women, and 0.63 (95% CI, 0.53‐0.73) for men. The results did not change substantively when the analysis was restricted to adenomatous lesions or to screening procedures without any additional incidental indication.CONCLUSIONS: The NCI CRC risk‐assessment tool displays modest discriminatory accuracy in predicting AN at screening colonoscopy in a diverse population. This tool may aid shared decision‐making in clinical practice. Cancer 2016. © 2016 American Cancer Society. Cancer 2016;122:2663–2670. © 2016 American Cancer Society.Keywords
Funding Information
- Stanford Cancer Institute Innovation Award
This publication has 41 references indexed in Scilit:
- Effect of Flexible Sigmoidoscopy-Based Screening on Incidence and Mortality of Colorectal Cancer: A Systematic Review and Meta-Analysis of Randomized Controlled TrialsPLOS Medicine, 2012
- Colonoscopic Polypectomy and Long-Term Prevention of Colorectal-Cancer DeathsNew England Journal of Medicine, 2012
- Derivation and Validation of a Prediction Rule for Estimating Advanced Colorectal Neoplasm Risk in Average-Risk ChineseAmerican Journal of Epidemiology, 2012
- Communication of Uncertainty Regarding Individualized Cancer Risk EstimatesMedical Decision Making, 2010
- Population-Based Family History–Specific Risks for Colorectal Cancer: A Constellation ApproachGastroenterology, 2010
- Cumulative Risk of Colon Cancer up to Age 70 Years by Risk Factor Status Using Data From the Nurses' Health StudyAmerican Journal of Epidemiology, 2009
- Development and Validation of a Colon Cancer Risk Assessment Tool for Patients Undergoing ColonoscopyAmerican Journal of Gastroenterology, 2009
- Validation of a Colorectal Cancer Risk Prediction Model Among White Patients Age 50 Years and OlderJournal of Clinical Oncology, 2009
- Colorectal Cancer Risk Prediction Tool for White Men and Women Without Known SusceptibilityJournal of Clinical Oncology, 2009
- Screening and Surveillance for the Early Detection of Colorectal Cancer and Adenomatous Polyps, 2008: A Joint Guideline From the American Cancer Society, the US Multi-Society Task Force on Colorectal Cancer, and the American College of RadiologyGastroenterology, 2008