Yang Xin, Ding Shengjie, Guo Jinlu, Yang Jingze, Du Fan, Liu Shi
Department of Gastroenterology, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, 430022, China.
Sci Rep. 2025 Aug 26;15(1):31509. doi: 10.1038/s41598-025-17421-3.
This study aimed to develop and validate a practical nomogram for differentiating between benign and malignant pancreatic masses. A total of 494 patients with pancreatic mass lesions, confirmed by histopathology, were enrolled from Wuhan Union Medical College Hospital between January 2020 and May 2022. The participants were randomly divided into development and validation groups in a 7:3 ratio. Using multivariate logistic regression, the nomogram was constructed based on five independent predictors: blood type, CA19-9, IgG4, anorexia, and weight loss. The model's performance was assessed using receiver operating characteristic (ROC) curve analysis and calibration curves. In the development and validation sets, the areas under the ROC curve were 0.932 and 0.957, respectively. The nomogram demonstrated a high net benefit in the clinical decision curve analysis. Based on the model, pancreatic malignancy risk was classified as low (< 4%), moderate (4%-71%), and high (> 71%). This nomogram provides an easy-to-use, efficient tool for the early differentiation of pancreatic malignancies and could be implemented in primary, secondary, and emergency care settings to facilitate the timely referral of patients to higher-level hospitals.
本研究旨在开发并验证一种用于区分胰腺良性和恶性肿块的实用列线图。2020年1月至2022年5月期间，从武汉协和医院招募了494例经组织病理学确诊的胰腺肿块病变患者。参与者按7:3的比例随机分为开发组和验证组。使用多因素逻辑回归，基于血型、CA19-9、IgG4、厌食和体重减轻这五个独立预测因素构建列线图。使用受试者操作特征（ROC）曲线分析和校准曲线评估模型性能。在开发集和验证集中，ROC曲线下面积分别为0.932和0.957。列线图在临床决策曲线分析中显示出较高的净效益。基于该模型，胰腺恶性肿瘤风险分为低（<4%）、中（4%-71%）和高（>71%）。这种列线图为胰腺恶性肿瘤的早期鉴别提供了一种易于使用、高效的工具，可在基层、二级和急诊医疗环境中实施，以促进患者及时转诊至上级医院。
