Work

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

Year: 2026

Type: article

Abstract: Objectives: To develop and validate a model for predicting the risk of pathological upgrading in patients with colorectal polyps. Methods: = 178) at a ratio of 7:3. In the training cohort, least absolute shrinkage and selection operator (LASSO) regression was used to select possible predictive factors. Multivariable logistic regression was then applied to identify independent risk factors. A nomogram was developed to show the prediction model in a visual way. The performance of the model was assessed using the receiver operating characteristic (ROC) curve, calibration plot, Hosmer-Lemeshow goodness-of-fit test, and decision curve analysis (DCA). SHapley Additive Explanations (SHAP) were also used to help explain the model results. Results: The polyp was located in the rectum, with an MTD ≥ 30 mm. The polyp had a villous structure. Erosion of the polyp and redness of the polyp surface were identified as significant predictors of pathological escalation in patients with colorectal polyps. A nomogram developed based on these predictors showed excellent predictive performance. The area under the ROC curve (AUC) for the training set and the test set is 0.890 and 0.922, respectively. The calibration curve and the Hosmer-Lemeshow test show a high degree of consistency between the predicted and observed results, and DCA confirms that the model has superior clinical practicality. Conclusion: This study developed and validated a risk prediction model for pathological upgrade of colorectal polyps based on five endoscopic factors, including rectal location, maximum tumor diameter (MTD) ≥ 30 mm, villous structure, erosion, and a red surface color. The model serves as a practical clinical tool that allows endoscopists to assess patient risk with high accuracy before treatment. By helping identify high-risk polyps that may need wider resection or closer follow-up, the model supports more personalized treatment decisions and may reduce both under-treatment and over-treatment. Its use is expected to improve individual patient management and enhance the effectiveness of colorectal cancer prevention.

Source: Frontiers in Medicine

DOI: 10.3389/fmed.2026.1748424

Authors: Ziyao Cheng, Chang Zhang, Feng Yu

Institutions: First Affiliated Hospital of Nanchang University, Fourth Affiliated Hospital of Anhui Medical University

Language: English

Topic: Colorectal Cancer Screening and Detection

Subfield: Oncology

Field: Medicine

Domain: Health Sciences

SDG: Good health and well-being

Open Access status: gold

Estimated APC paid: $3,801