Risk Stratification to Optimize Colorectal Cancer Screening: Development and Validation of a Decision-Tree Model for...

Fecal immunochemical test-based colorectal cancer screening is effective; however, colonoscopy demand exceeds endoscopy capacity. Delays in colonoscopy after a positive test reduce screening effectiveness. We developed and validated a decision-tree model to stratify colorectal cancer risk among individuals with a positive test to prioritize colonoscopy for those at highest risk.

Prospective study including participants with a positive fecal immunochemical test referred for colonoscopy within a Spanish regional population-based screening program (2021-2024). Individuals were randomly assigned to derivation (65%) and validation (35%) cohorts using a stratified random split. Fecal hemoglobin concentration, clinical and laboratory data were collected to develop the decision-tree model. The primary outcomes were colorectal cancer and advanced-stage colorectal cancer.

A total of 1773 participants were included. Colorectal cancer was detected in 97 participants (5.5%), including 25 cases (1.5%) with advanced-stage disease. The model stratified participants into low-, intermediate-, high-, and very high-risk groups. A progressive increase in the prevalence of colorectal cancer and advanced-stage disease was observed across risk groups in both cohorts (p < 0.001). The very high-risk group (fecal hemoglobin > 100 μg/g and > 65 years) included 6.7% of participants, with colorectal cancer and advanced-stage prevalence of 21.8% and 7.6%, respectively. Prioritizing colonoscopy for this group yielded a number needed to scope of 4.6 for colorectal cancer and 13.2 for advanced-stage disease, both significantly lower than the current one-size-fits-all strategy.

A decision-tree model was developed to stratify the risk of colorectal cancer and optimize endoscopic resources in screening programs.