Tenofovir Versus Entecavir for Hepatocellular Carcinoma Prevention in an International Consortium of Chronic Hepatitis B

作者：Yao‐Chun Hsu, Grace Lai‐Hung Wong, Chien‐Hung Chen, Cheng‐Yuan Peng, Ming‐Lun Yeh, Ka Shing Cheung, Hidenori Toyoda, Chung‐Feng Huang, Huy N. Trinh, Qing Xie, Masaru Enomoto, Li Liu, Satoshi Yasuda, Yasuhito Tanaka, Ritsuzo Kozuka, Pei‐Chien Tsai, Yen‐Tsung Huang, Christopher K. Wong, Rui Huang, Tyng‐Yuan Jang, Joseph Hoang, Hwai‐I Yang, Jiayi Li, Dong-Hyun Lee, Hirokazu Takahashi, Jian Q. Zhang, Eiichi Ogawa, Changqing Zhao, Chenghai Liu, Norihiro Furusyo, Yuichiro Eguchi, Clifford Wong, Chao Wu, Takashi Kumada, Man‐Fung Yuen, Ming‐Lung Yu, Mindie H. Nguyen · 发表于：The American Journal of Gastroenterology · 年份：2019 · DOI：10.14309/ajg.0000000000000428 · 被引用次数：102 · 研究领域：Hepatitis B Virus Studies、Hepatitis C virus research、Hepatocellular Carcinoma Treatment and Prognosis

INTRODUCTION: It is unclear whether entecavir (ETV) and tenofovir disoproxil fumarate (TDF) differ in their effectiveness for preventing hepatocellular carcinoma (HCC) in patients with chronic hepatitis B (CHB). METHODS: This retrospective cohort study analyzed an international consortium that encompassed 19 centers from 6 countries or regions composed of previously untreated CHB patients then treated with either ETV or TDF monotherapy. Those who developed HCC before antiviral treatment or within 1 year of therapy were excluded. The association between antiviral regimen and HCC risk was evaluated using competing-risk survival regression. We also applied propensity score matching (PSM) to 1:1 balance the 2 treatment cohorts. A total of 5,537 patients were eligible (n = 4,837 received ETV and n = 700 received TDF) and observed for HCC occurrence until December 23, 2018. Before PSM, the TDF cohort was significantly younger and had generally less advanced diseases. RESULTS: In the unadjusted analysis, TDF was associated with a lower risk of HCC (subdistribution hazard ratio [SHR], 0.45; 95% confidence interval [CI], 0.26-0.79; P = 0.005). The multivariable analysis, however, found that the association between TDF and HCC no longer existed (SHR, 0.81; 95% CI, 0.42-1.56; P = 0.52) after adjustment for age, sex, country, albumin, platelet, α-fetoprotein, cirrhosis, and diabetes mellitus. Furthermore, the PSM analysis (n = 1,040) found no between-cohort differences in HCC incidences ...