Abstract

Objective: To compare microvascular invasion (McVI) with parameters defined by the Milan criteria in predicting tumor recurrence and overall survival (OS) in patients with surgical resection (SR) for hepatocellular carcinoma (HCC). Summary Background Data: Although the Milan criteria is discriminatory for selecting patients with good outcomes in liver transplantation and SR for HCC, it neither adequately predict tumor recurrence nor explain differences in survival for patients with good liver function. McVI is a strong indicator of intrahepatic metastasis in HCC, but its relative significance for predicting clinical outcomes compared to the Milan criteria is unclear. Methods: Patients undergoing SR with curative intent from January 2000 to March 2009 at the Singapore General Hospital were followed up for long-term outcomes till January 1, 2010. They were stratified first by the Milan criteria and then by the presence of McVI and compared relative to OS. Results: Altogether, 454 of the 515 patients received curative SR. There were stratified into 4 groups (Milan+, McVI−), (Milan+, McVI+), (Milan−, McVI−), and (Milan−, McVI+). All pair-wise comparisons between groups relative to OS were significant except (Milan+, McVI−) (OS, 90%, 73%, and 60% at 1, 3, and 5 years) with (Milan−, McVI−) (OS, 86%, 71%, and 61% at 1, 3, 5 years) and (Milan+, McVI+) with (Milan−, McVI+). Multivariate Cox regression analysis showed that McVI was predictive of OS, after which Milan status did not add additional discriminative information. Conclusions: McVI is a better predictor of tumor recurrence and OS than the Milan criteria after SR for HCC. Assessment of McVI should aid in patient selection for adjuvant treatments to improve outcomes after SR.