LI-RADS categories and clinicopathological features predict recurrence-free survival in patients with radical...

To assess the value of LI-RADS categories and clinicopathological characteristics in predicting recurrence-free survival (RFS) after radical resection of hepatocellular carcinoma (HCC).

This retrospective study included HCC patients who underwent surgical resection between January 2014 and July 2018. Two radiologists independently evaluated the LI-RADS v2018 categories. Clinicopathological parameters were documented. Univariate and multivariate Cox regression analyses were performed to identify independent risk factors for RFS. These risk factors were then used to constructe a nomogram, its performance was evaluated with receiver operating characteristic (ROC) curve analysis. Internal validation was performed via bootstrap resampling, with calibration curves and the C-index assessing model consistency.

A total of 218 patients (mean age: 49.47 ± 11.14 years; 184 males) were included. Multivariate analysis revealed that tumor maximum diameter (hazard ratio (HR) = 1.01 [95 % confidence interval (CI): 1.01, 1.01], p < 0.001), tumor number (>1; HR = 1.95 [95 % CI: 1.39, 2.74], p < 0.001), LI-RADS category ((LR-TIV: HR = 2.01 [95 % CI: 1.33, 3.04], p < 0.001), and MVI grading (M2: HR = 2.70 [95 % CI: 1.70, 4.31], p < 0.001) were independent risk factors and were used to construct a nomogram. The areas under the ROC curves (AUCs) in predicting the 1-, 2-, and 5-year RFS rates were 0.84 (95 % CI: 0.79, 0.89), 0.77 (95 % CI: 0.71, 0.83) and 0.77 (95 % CI: 0.69, 0.84), respectively. The bootstrap-corrected C-index was 0.73 (95 % CI: 0.69, 0.77).

The nomogram combining LI-RADS category, tumor maximum diameter and number, and MVI grading could provide a practical tool for post-resection RFS prediction in HCC patients.