Intraductal papillary mucinous neoplasm (IPMN) is a relatively uncommon pancreatic cystic tumor that serves as a significant precursor lesion closely associated with pancreatic cancer. This study aimed to identify risk factors for malignant transformation of IPMN and develop a corresponding predictive model.

Patients with IPMN admitted to Shanghai General Hospital between May 2012 and January 2025 were included in this study. Patients were stratified into benign and malignant groups. Characteristic parameters and laboratory results were collected. The training set and test set were randomly divided at a ratio of 7:3. Least absolute shrinkage and selection operator regression was used to select potential prognostic factors. A nomogram was developed by firth penalty logistic regression. Receiver operating characteristic curves and calibration curves were used to evaluate the model’s predictive performance.

Among 121 patients analysed, four independent predictors of malignancy were identified: CA19-9 Elevation, classified as MD/MT, MTD_Log, and MPD_bc. We developed a nomogram with an area under the curve of 0.870 in the training set and 0.808 in the test set. The model showed strong predictive ability and showed good clinical usefulness.

CA19-9 Elevation, classified as MD/MT, MTD_Log, and MPD_bc are independent risk factors for malignant IPMN. The established prediction model demonstrates good accuracy and may provide valuable guidance for clinical decision-making.