Hepatocellular carcinoma incidence with tenofovir versus entecavir in chronic hepatitis B: a systematic review and meta-analysis

Background It is unclear whether tenofovir disoproxil fumarate and entecavir differ in their association with risk of hepatocellular carcinoma in patients with chronic hepatitis B, and previous meta-analyses have shown conflicting conclusions with substantial heterogeneity. We aimed to analyse the updated data and elucidate the source of heterogeneity. Methods We searched PubMed, Embase, Web of Science, and the Cochrane library for relevant studies with time-toevent data for incident hepatocellular carcinoma occurring in patients with chronic hepatitis B who received tenofovir disoproxil fumarate or entecavir monotherapy with follow-up of at least 1 year. Studies published between Jan 1, 2006, and April 17, 2020, and abstracts from international conferences in 2018 and 2019 were included. We pooled covariate adjusted hazard ratios (HRs) for hepatocellular carcinoma using a random-effects model, assessed heterogeneity among included studies using the I-2 statistic and Cochran's Q test, and identified the source of heterogeneity using prespecified subgroup analyses. This study is registered with PROSPERO, ID CRD42020176513. Findings 31 studies involving 119 053 patients were analysed. The 5-year cumulative incidence of hepatocellular carcinoma was 5.97% (95% CI 5.81-6.13, 28 studies) for entecavir and 3.06% (2.86-3.26, 13 studies) for tenofovir disoproxil fumarate in studies with unmatched populations (p<0.0001). For all eight studies matched by propensity score, the 5-year cumulative incidence was 3.44% (95% CI 3.08-3.80) for entecavir and 3.39% (2.94-3.83) for tenofovir disoproxil fumarate (p=0.87). Analysis of 14 comparative studies with covariate adjustment found that tenofovir disoproxil fumarate and entecavir had similar risk of hepatocellular carcinoma (primary outcome); adjusted HR 0.88, 95% CI 0.73-1.07; p=0.20), although heterogeneity was significant (I-2=56.4%, p=0.0038). In a subgroup analysis for hospital-based clinical cohorts, there was no difference in hepatocellular carcinoma incidence between the two regimens (adjusted HR 1.03, 95% CI 0.88-1.21; I-2=0%). However, tenofovir disoproxil fumarate was associated with a lower risk of hepatocellular carcinoma compared with entecavir in administrative database research (adjusted HR 0.67, 0.59-0.76; I-2=0%). Interpretation Our study found no significant difference between tenofovir disoproxil fumarate and entecavir in their association with incident hepatocellular carcinoma. We suggest that treatment should be guided by patient tolerability and affordability rather than whether one drug is more effective than the other.