Yang Dujiang, Liu Xijiao, Li Mao, Li Zhenlu, Ke Nengwen, Xiong Junjie
Division of Pancreatic Surgery, Department of General Surgery, West China Hospital, Sichuan University, Chengdu, China.
Department of Radiology, West China Hospital, Sichuan University, Chengdu, China.
Cancer Med. 2025 Sep;14(17):e71182. doi: 10.1002/cam4.71182.
Pancreatic intraductal papillary mucinous neoplasms (IPMN) are precursors to pancreatic cancer, with an increasing incidence due to advances in imaging techniques. This study aimed to identify risk factors for malignant transformation in IPMN and develop a predictive model using data from a large medical center in western China.
Patients with IPMN admitted to West China Hospital between January 2010 and February 2022 were included in this study. They were divided into benign and malignant. 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 logistic regression. Receiver operating characteristic curves and calibration curves were used to evaluate the model's predictive performance.
We retrospectively analyzed 182 patients, identifying six independent predictors of malignancy: classification, cyst wall thickening, abrupt changes in main pancreatic duct caliber, maximum tumor diameter, maximum main pancreatic duct diameter, and lnCA19-9. We developed a nomogram with an area under the curve of 0.86 in the training set and 0.81 in the test set. The model showed strong predictive ability, providing a valuable tool for clinicians to guide preoperative decision-making.
Our study offers the first predictive model for malignant IPMN in western China and highlights the importance of comprehensive risk assessment, incorporating clinical, imaging, and laboratory data.
胰腺导管内乳头状黏液性肿瘤（IPMN）是胰腺癌的癌前病变，随着成像技术的进步，其发病率不断上升。本研究旨在确定IPMN恶变的危险因素，并利用中国西部一家大型医疗中心的数据建立预测模型。
本研究纳入了2010年1月至2022年2月期间入住华西医院的IPMN患者。将他们分为良性和恶性两组。收集特征参数和实验室检查结果。训练集和测试集按7:3的比例随机划分。采用最小绝对收缩和选择算子回归来选择潜在的预后因素。通过逻辑回归建立列线图。采用受试者工作特征曲线和校准曲线来评估模型的预测性能。
我们回顾性分析了182例患者，确定了六个恶性独立预测因素：分类、囊壁增厚、主胰管口径突然改变、最大肿瘤直径、最大主胰管直径和lnCA19-9。我们建立了一个列线图，训练集曲线下面积为0.86，测试集为0.81。该模型显示出强大的预测能力，为临床医生指导术前决策提供了一个有价值的工具。
我们的研究提供了中国西部首个IPMN恶变预测模型，并强调了综合风险评估的重要性，包括临床、影像和实验室数据。
