AbstractBACKGROUND & AIMS:
We performed a propensity-score matched analysis to investigate whether entecavir, compared with lamivudine, can reduce risk of hepatocellular carcinoma (HCC) in patients with chronic hepatitis B after adjusting for level of fibrosis.METHODS:
We performed a retrospective study of 1079 patients with chronic hepatitis B who received first-line therapy with lamivudine (n = 435) or entecavir (n = 644) from 2006 through 2013. Only patients with available liver stiffness value measured by transient elastography were recruited. Liver cirrhosis was diagnosed by ultrasonography. To adjust for the imbalance of patients treated with lamivudine versus entecavir, we performed propensity-score matching (PSM), at a ratio of 1:1, using 7 factors (age, sex, hepatitis B e antigen, alanine aminotransferase, serum albumin, platelet count, and liver stiffness; PSM1) or 8 factors (variables of PSM1 plus ultrasonography measurements of cirrhosis; PSM2). Patients with virologic breakthrough or resistance mutations received rescue therapy.RESULTS:
Over the 7-year period, 91 patients developed HCC and 104 had liver-related events in the entire cohort. In multivariate analyses, level of fibrosis, but not antiviral regimen, was independently associated with risk of HCC (P< .05). The PSM1 group included 342 pairs of patients and the PSM2 group included 338 pairs. Similar proportions of patients given lamivudine versus entecavir developed HCC in each model (10.5% given lamivudine vs 9.9% given entecavir in PSM1 and 11.9% vs 12.6% in PSM2; allP> .05). When PSM was applied to patients with liver stiffness value ≤13 kPa or >13 kPa, patients given lamivudine versus entecavir still had similar cumulative rates of HCC development (allP> .05).CONCLUSIONS:
In a PSM analysis, we associated level of fibrosis, rather than antiviral regimen, with risk of HCC, when patients received appropriate rescue therapy in case of virologic breakthrough or resistance mutations.