Establishment and validation of a nomogram for postoperative disease progression in patients with primary liver cancer
10.3760/cma.j.cn113884-20250108-00006
- VernacularTitle:原发性肝癌患者术后疾病进展列线图模型的建立与验证
- Author:
Tianchen XU
1
;
Ru JIA
;
Ruiqi ZHANG
;
Yuling WANG
;
Xuelian CHEN
Author Information
1. 江苏大学附属昆山医院放射科，昆山 215300
- Publication Type:Journal Article
- Keywords:
Liver neoplasms;
Quantitative CT;
Muscle;
Fat;
Prediction model
- From:
Chinese Journal of Hepatobiliary Surgery
2025;31(4):247-252
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To establish and validate a nomogram for postoperative disease progression (including recurrence, metastasis, and death) in patients with primary liver cancer (PLC) based on quantitative CT measurements of relevant indicators.Methods:Clinical data of 290 patients with PLC admitted to Zhongshan Hospital Affiliated to Fudan University and Kunshan Hospital Affiliated to Jiangsu University from January 2016 to December 2021 were retrospectively collected, including 177 males and 113 females, aged (60.3±11.9) years. Two hundred and three patients admitted to Zhongshan Hospital Affiliated to Fudan University were used as the training set, and 87 patients admitted to Kunshan Hospital Affiliated to Jiangsu University were used as the validation set. The patient's condition of ascites , tumor length, number of lesions, tumor differentiation degree, relevant indicators of quantitative CT detection (including decreased muscle mass and increased intra-abdominal fat area), prognosis and other clinical data were recorded. The influencing factors of postoperative disease progression was analyzed through multiple logistic regression in the training set, and the nomogram model was constructed based on the results of multiple factor analysis. The predictive performance of the model was evaluated using receiver operating characteristic (ROC) curves and calibration curves. The clinical applicability of predictive models was evaluated using the decision curve analysis.Results:The results of multiple logistic regression analysis showed that the increase in maximum tumor diameter ( OR=1.519, 95% CI: 1.251-1.843), multiple lesions ( OR=3.193, 95% CI: 1.493-6.830), low tumor differentiation ( OR=5.604, 95% CI: 2.442-12.863), ascites ( OR=3.321, 95% CI: 1.166-9.463), portal vein tumor thrombus ( OR=3.990, 95% CI: 1.681-9.474), decreased muscle mass ( OR=2.173, 95% CI: 1.051-4.492) and increased intra-abdominal fat area ( OR=2.634, 95% CI: 1.276-5.438) were independent risk factors for postoperative disease progression in patients with PLC (all P<0.05). A nomogram was constructed based on the above variables, and the area under the ROC curve for predicting postoperative disease progression in patients with PLC in the training set and validation set was 0.862 (95% CI: 0.810-0.914) and 0.879 (95% CI: 0.806-0.953), respectively. The calibration curve and ideal curve fit well, indicating that the predicted situation was basically consistent with the actual situation. Decision curve analysis showed that the column chart model had a high clinical net benefit and good clinical prediction effectiveness. Conclusion:The nomogram constructed based on the maximum diameter of the tumor, the number of lesions, the degree of tumor differentiation, ascites, portal vein tumor thrombus, decreased muscle mass, and increased intra-abdominal fat area has good predictive power for postoperative disease progression in patients with PLC.
