A Pathological Prediction Model and Scoring System Including Endoscopic Signs for Duodenal Papilla Neoplasms: A...

Area of Science:

• Gastroenterology
• Diagnostic Imaging
• Oncology

Background:

• Duodenal papilla neoplasms (DPNs) diagnosis requires effective differentiation between benign and malignant types.
• The diagnostic value of endoscopic ultrasound (EUS) for DPNs is not fully established.
• Accurate pathological prediction is crucial for appropriate DPN management.

Purpose of the Study:

• To evaluate the role of EUS and other clinical indicators in differentiating benign from malignant DPNs.
• To develop and validate a predictive model for the pathological diagnosis of DPNs.
• To assess the diagnostic performance of EUS in DPNs.

Main Methods:

• Retrospective collection of clinical and imaging data from DPNs patients.
• Application of Least Absolute Shrinkage and Selection Operator (LASSO) regression for predictor screening.
• Development and validation of a nomogram and scoring system using training and test cohorts.

Main Results:

• Identified age, MRI, EUS size, echo intensity, and papilla appearance as independent predictors.
• The nomogram achieved a C-index of 0.876 with AUCs of 0.88 (training) and 0.82 (test).
• A scoring system with an optimal cutoff of 14 points demonstrated good performance.

Conclusions:

• Age, MRI, EUS size, and papilla appearance are significant risk factors for malignant DPNs.
• EUS shows potential for significant contributions to the differential diagnosis of DPNs.
• The developed nomogram and scoring system offer a valuable tool for pathological prediction in DPNs.