The aim of the study was to compare SURG vs SOR regarding the OS and progression-free survival (PFS) in a real-world clinical scenario.
The treatment for advanced nonmetastatic HCC belonging to the Barcelona Clinic Liver Cancer stage C (BCLC C) is still controversial.
BCLC C patients without extrahepatic spread and tumoral invasion of the main portal trunk were considered. Surgical patients were obtained from the HE.RC.O.LE.S. Register, whereas sorafenib patients were obtained from the ITA.LI.CA register The inverse probability weighting (IPW) method was adopted to balance the confounders between the 2 groups.
Between 2008 and 2019, 478 patients were enrolled: 303 in SURG and 175 in SOR group. Eastern Cooperative Oncological Group Performance Status (ECOG-PS), presence of cirrhosis, steatosis, Child-Pugh grade, hepatitis B virus and hepatitis C virus, alcohol intake, collateral veins, bilobar disease, localization of the tumor thrombus, number of nodules, alpha-fetoprotein, age, and Charlson Comorbidity index were weighted by IPW to create two balanced pseudo-populations: SURG = 374 and SOR = 263. After IPW, 1–3–5 years OS was 83.6%, 68.1%, 55.9% for SURG, and 42.3%, 17.8%, 12.8% for SOR (P < 0.001). Similar trends were observed after subgrouping patients by ECOG-PS = 0 and ECOG-PS >0, and by the intrahepatic location of portal vein invasion. At Cox regression, sorafenib treatment (hazard ratio 4.436; 95% confidence interval 3.19–6.15; P < 0.001) and Charlson Index (hazard ratio 1.162; 95% confidence interval 1.06–1.27; P = 0.010) were the only independent predictors of mortality. PFS at 1–3–5 years were 65.9%, 40.3%, 24.3% for SURG and 21.6%, 3.5%, 2.9% for SOR (P = 0.007).
In BCLC C patients without extrahepatic spread but with intrahepatic portal invasion, liver resection, if feasible, was followed by better OS and PFS compared with sorafenib.