Development and validation of a predictive model for pathological upgrading in colorectal polyps based on endoscopic...

Area of Science:

• Gastroenterology
• Oncology
• Medical Diagnostics

Background:

• Colorectal polyps require accurate risk assessment for pathological upgrading.
• Endoscopic features can indicate the risk of malignancy.
• Predictive models can improve clinical decision-making in polyp management.

Purpose of the Study:

• To develop and validate a predictive model for pathological upgrading in colorectal polyps.
• To identify key endoscopic factors associated with pathological escalation.
• To provide a tool for endoscopists to assess patient risk accurately.

Main Methods:

• Prospective study of 593 patients with colorectal polyps.
• Utilized least absolute shrinkage and selection operator (LASSO) regression and multivariable logistic regression.
• Developed a nomogram for visual risk prediction, validated with ROC curves, calibration plots, and decision curve analysis (DCA).

Main Results:

• Rectal location, maximum tumor diameter (MTD) ≥ 30 mm, villous structure, erosion, and red surface color were significant predictors.
• The nomogram demonstrated excellent predictive performance with AUCs of 0.890 (training) and 0.922 (testing).
• High consistency between predicted and observed results, with superior clinical practicality confirmed by DCA.

Conclusions:

• A validated risk prediction model based on five endoscopic factors for colorectal polyp pathological upgrading was developed.
• The model serves as a practical clinical tool for accurate risk assessment before treatment.
• This tool supports personalized treatment decisions, potentially reducing under-treatment and over-treatment, and enhancing colorectal cancer prevention.