Identification of patients with favorable prognosis after resection in intermediate-stage-hepatocellular carcinoma.

BACKGROUNDS: It is unclear which patients benefit from resection in intermediate-stage-hepatocellular carcinoma (HCC). The authors aimed to identify high-risk patients for early recurrence among patients with resectable intermediate-stage HCC. METHODS: This multicenter retrospective study included patients who underwent resection or trans-arterial chemoembolization (TACE) for intermediate-stage HCC (2008-2019). Multivariable Cox proportional analysis was performed to identify high-risk patients when treated with resection. A prediction score for 2-year recurrence-free survival (RFS) was developed using the training cohort and validated. The 2-year RFS in each risk group was compared with that in TACE group, after propensity score matching (PSM). RESULTS: A total of 1686 patients were included (480 and 1206 patients in the resection and TACE groups). During a median follow-up of 31.4 months, the 2-year RFS was significantly higher in the resection (47.7%) than in the TACE group (19.8%) [adjusted hazard ratio (aHR)=1.471, 95% CI: 1.199-1.803, P <0.001). On multivariate analysis, alpha-fetoprotein ≥5.0 ng/ml (aHR=0.202), ALBI grade ≥2 (aHR=0.709), tumor number ≥3 (aHR=0.404), and maximal tumor size ≥5 cm (aHR=0.323) were significantly associated with the lower risk of 2-year RFS in the resection group. The newly developed Surgery Risk score in BCLC-B (SR-B score) with four significant risk factors showed an area under the curve of 0.801 for the 2-year RFS and was validated. Based on the SR-B score, low-risk patients had a significantly higher 2-year RFS (training: aHR=5.834; validation: aHR=5.675) than high-risk patients (all P <0.001) did. In a PSM cohort, a low-risk resection group had a significantly higher (aHR=3.891); a high-risk resection group had a comparable 2-year RFS to those treated with TACE (aHR=0.816). CONCLUSIONS: Resection may be beneficial for resectable intermediate-stage HCC based on the SR-B score.