Preoperative MR imaging for predicting early recurrence of solitary hepatocellular carcinoma without microvascular invasion

Objectives: This study aimed to identify preoperative MR imaging features for predicting early recurrence after curative resection of solitary hepatocellular carcinoma (HCC) without microvascular invasion (MVI). Methods: 124 patients with MVI-negative HCC who underwent preoperative dynamic contrast-enhanced 1.5-T MR imaging before surgical resection were included. Liver Imaging Reporting and Data System (LI-RADS v2018) imaging features and three non-LI-RADS MR imaging features for predicting early recurrence (intrahepatic recurrence<2 years) were identified by univariable and multivariable analyses. A nomogram was constructed for individualized risk estimation, and its predictive accuracy and discriminative ability were identified by concordance index (C-index) and calibration curve. Results: In multivariable analysis, tumor size (p = 0.045), nonsmooth tumor margin (p = 0.013), and presence of mosaic architecture (p = 0.035) were independent significant variables associated with early recurrence. These were all incorporated to establish the nomogram. The C-index of the nomogram was 0.743 (95 % CI: 0.697-0.788). Conclusion: At dynamic contrast-enhanced MR imaging, tumor size, nonsmooth tumor margin, and presence of mosaic architecture may be helpful to predict early recurrence of solitary HCC without MVI after curative resection.