Preoperative Prediction of Microvascular Invasion with Gadoxetic Acid-Enhanced Magnetic Resonance Imaging in Patients with Single Hepatocellular Carcinoma: The Implication of Surgical Decision on the Extent of Liver Resection

Abstract

Introduction: Microvascular invasion (MVI) is one of the most important prognostic factors for hepatocellular carcinoma (HCC) recurrence, but its application in preoperative clinical decisions is limited. This study aimed to identify preoperative predictive factors for MVI in HCC and further evaluate oncologic outcomes of different types and extents of hepatectomy according to stratified risk of MVI.

Methods: Patients with surgically resected single HCC (≤5 cm) who underwent preoperative gadoxetic acid-enhanced magnetic resonance imaging (MRI) were included in a single-center retrospective study. Two radiologists reviewed the images with no clinical, pathological, or prognostic information. Significant predictive factors for MVI were identified using logistic regression analysis against pathologic MVI and used to stratify patients. In the subgroup analysis, long-term outcomes of the stratified patients were analyzed using the Kaplan-Meier method with log-rank test and compared between anatomical and nonanatomical or major and minor resection.

Results: A total of 408 patients, 318 men and 90 women, with a mean age of 56.7 years were included. Elevated levels of tumor markers (alpha-fetoprotein [α-FP] ≥25 ng/mL and PIVKA-II ≥40 mAU/mL) and three MRI features (tumor size ≥3 cm, non-smooth tumor margin, and arterial peritumoral enhancement) were independent predictive factors for MVI. As the MVI risk increased from low (no predictive factor) and intermediate (1-2 factors) to high-risk (3-4 factors), recurrence-free and overall survival of each group significantly decreased (p = 0.001). In the high MVI risk group, 5-year cumulative recurrence rate was significantly lower in patients who underwent major compared to minor hepatectomy (26.6 vs. 59.8%, p = 0.027).

Conclusion: Tumor markers and MRI features can predict the risk of MVI and prognosis after hepatectomy. Patients with high MVI risk had the worst prognosis among the three groups, and major hepatectomy improved long-term outcomes in these high-risk patients.

Keywords: Hepatectomy; Hepatocellular carcinoma; Magnetic resonance imaging; Microscopic vascular invasion; Recurrence.

Fig. 1.

A flow diagram of patient selection.

Fig. 2.

a, b Predicted risk of microscopic vascular invasion and (c) ROC according to the number of predictive factors for MVI. As the number of predictive factors increased, the risk of MVI increased, specifically, microvessel and microscopic portal vein invasion increased (p < 0.001). The AUC, sensitivity, and specificity were 0.706 (95% confidence interval: 0.655–0.757), 64.2%, and 66.4%, respectively.

Fig. 3.

a, b Overall survival and recurrence-free survival depending on MVI risk OS and RFS gradually got worse as the MVI risk increased and were worst in the high-risk group (OS: p = 0.004; RFS: p < 0.001.

Fig. 4.

Overall survival and cumulative recurrence rate according to extent of hepatectomy on low (a, b), intermediate (c, d), and high MVI risk group (e, f), respectively. Cumulative recurrence rate was similar between major versus minor hepatectomy in low (10.5% vs. 15.7% in 5-year, p = 0.574) and intermediate (16.9% vs. 33.1% in 5-year, p = 0.078) groups, and significantly worse in the high-risk group (26.6% vs. 59.8%, p = 0.040) by the Kaplan-Meier method and log-rank test.