Background & aims: Sorafenib (SOR) is the standard of care for patients with hepatocellular carcinoma (HCC) and portal vein invasion (PVI), based on the results of phase 3 trials. However, radioembolization (RE) using yttrium-90 microspheres has been shown to achieve higher response rates and better survival in large cohorts and phase 2 trials. This study aimed to compare survival of HCC patients with PVI treated by RE or SOR.
Methods: Survival among patients with HCC and PVI treated with RE or SOR in four Spanish hospitals between 2005 and 2013 was analysed retrospectively. Kaplan-Meier survival curves were plotted and baseline variables tested for prognostic value using the log-rank test. A multivariate prognostic model including variables identified in the univariate analysis and adjusted by a propensity score based on factors that may determine the probability of exposure to RE was generated using Cox regression analyses.
Results: After a median follow-up of 6 months, 60 deaths had occurred: 38 and 22 in SOR and RE groups respectively. Median survival was 6.7 months (95%CI 5.2-8.1 months) for the entire cohort, and 8.8 months (95%CI 1.8-15.8) in the RE group and 5.4 months (95%CI 2.7-8.1) in the SOR group (P = 0.047). The difference in survival was still statistically significant when 13 patients in the RE group who started SOR after a median time of 8 months were censored from the analysis.
Conclusions: In a cohort of patients with HCC and PVI treatment with RE was associated with a more prolonged survival compared with SOR.
Keywords: hepatocellular carcinoma; portal vein invasion; radioembolization; selective internal radiation therapy; sorafenib.
© 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.