Development and validation of a simplified dynamic nomogram for preoperative prediction of microvascular invasion in hepatocellular carcinoma
Jianhao Zhang, Desheng Chen, Yutao Chen, Kang Yao, Xing Lyu, Linda Fan, Chenghao Zhao, Yongwei Hu, Kaiming He, Haobin Sun, Yufeng He, Wenjie Zheng, Yingcai ZhangAim
Microvascular invasion (MVI) is a key predictor of recurrence in hepatocellular carcinoma (HCC), but it is diagnosed pathologically after surgery. We aimed to develop and validate a simplified preoperative nomogram using routine markers and tumor burden score (TBS).
Methods
This retrospective study included 512 patients with HCC who underwent radical hepatectomy between June 2018 and 2023 (309 MVI-negative and 203 MVI-positive). Predictors were selected by univariate analysis and least absolute shrinkage and selection operator regression. Multivariable logistic regression was used to construct the model. Discrimination, calibration, and clinical utility were assessed using the area under the receiver operating characteristic curve (AUC), calibration curves, and decision curve analysis. An interactive dynamic nomogram was developed from the final model.
Results
The nomogram incorporated dichotomized aspartate aminotransferase (≥29.5 U/L), platelet count (≥127.5 × 10 9 /L), and alpha-fetoprotein (≥400 ng/ml), and TBS modeled as a continuous variable with restricted cubic splines. The model showed good discrimination (AUC = 0.720), good agreement between predicted and observed probabilities, and meaningful net benefit when the threshold probability exceeded 20%.
Conclusion
This simplified dynamic nomogram provides a practical noninvasive tool for preoperative MVI risk assessment and may support individualized decision-making in patients with HCC.