Development and Validation of a Nomogram Combining Hematological and Imaging Feature for Preoperative Prediction of Microvascular Invasion in Hepatocellular Carcinoma Patients

Background: Microvascular invasion (MVI) is a significant hazard factor which influences the recurrence and survival of hepatocellular carcinoma (HCC) patients after hepatectomy. We aimed to develop and validate a nomogram that combines hematological and imaging features of hepatocellular carcinoma patients for preoperative prediction of microvascular invasion. Methods: The study included a total of 709 HCC patients who underwent liver resection at the Liver Cancer Institute of Zhongshan Hospital, Fudan University (FDZS) between June 1,2015 and December 30,2016 (496 as training validation cohort and 213 as validation cohort). Blood biochemistry parameters and radiological features of dynamic contrast‑enhanced MRI were extracted. Lasso regression model and multivariable logistic regression were used for data dimension reduction, variable selection and development of the predicting model. The model was presented as a nomogram and its performance was assessed in terms of discrimination, calibration and clinical usefulness. Results: In the training cohort, 218 of 496 patients (43.9%) had histopathologically identified MVI compared to 88 of 213 patients (41.3%) in the validation cohort. Independent prognostic factors including alkaline phosphatase (ALP, >125 U/L), alpha-fetoprotein (AFP, within 20 - 400 or >400 ng/mL), protein induced by vitamin K absence-II (PVIKA-II, within 40 – 400 or >400 mAU/mL), tumor number, diameter, pseudo-capsule, tumor growth pattern and intratumor hemorrhage were incorporated in our nomogram. The model showed good discrimination and calibration , with a concordance indexes (C-index,0.82[ 95% CI: 0.7822-0.857]) in the training cohort and C-index (0.80, [95% CI: 0.772-0.837]) in the validation cohort. Decision curve analysis (DCA) also showed that this model was clinically useful. Interpretation: This study presents an optimal model that addresses and combines hematological and radiological features. By predicting the incidence of MVI preoperatively , this model could help surgeons choose the best treatment options for HCC patients before and after the operation. Funding Statement: This work was funded by grants from the Shanghai International Science and Technology Collaboration Program (18410721900), the National Natural Science Foundation of China (81472672), and Shanghai Municipal Key Clinical Specialty Declaration of Interests: All authors declare no potential conflicts of interest. Ethics Approval Statement: The study was approved by Ethics Committee of Zhongshan Hospital, Fudan University. Informed consent of clinical data used in this study was obtained from all patients without financial compensation.