Commentary on selected evidence

With advice from topic experts we selected 1 study for further commentary.

Research recommendation – Long-term safety of tenofovir disoproxil in chronic hepatitis B

We selected a randomised controlled trial by Buti et al. (2015) for a full commentary because this is a new piece of evidence that relates to one of the research recommendations from the original guideline.

What the guideline recommends

The research recommendation was to determine the long-term safety of tenofovir disoproxil, including the risk of clinically significant hypophosphataemia and related bone toxicity, in people with chronic hepatitis B. The cost effectiveness of routine monitoring for phosphate loss and bone disease in people with chronic hepatitis B who are receiving tenofovir disoproxil treatment was also considered to need further evaluation.


Study participants (n=641) included people who were hepatitis B e antigen (HBeAg)-positive or HBeAg-negative chronic hepatitis B (CHB), 18–69 years of age with compensated liver disease and Knodell necroinflammatory score C3. Participants were randomised to receive either tenofovir disoproxil fumarate (TDF) or adefovir dipivoxil (ADV) daily for 1 year, after which all participants received TDF for up to 7 years.


At study completion, 68% of the participants remained with the other 32% discontinuing treatment due to withdrawal of consent; loss to follow-up; investigator discretion; safety, tolerability, or efficacy reasons; HBsAg or HBeAg seroconversion; protocol violation; or sponsor decision. Among the participants who had abnormal serum alanine aminotransferase (ALT) at the beginning of the study, 80% of them experienced ALT normalisation by the end of year 7. Of the HBeAg-positive participants who remained on the study at year 7, 55% experienced HBeAg loss and 40% experienced seroconversion to anti-HBe indicating that TDF was both an effective and safe treatment for chronic hepatitis B.

Strengths and limitations


The study population matched those within the current guideline, also the methodology was fully and clearly reported along with the majority of the outcomes. Patient disposition details at year 7 were also reported in full, which aimed to add to the evidence base on the long term safety of treatment.


The study methodology described measuring several outcomes throughout the duration of the open label phase but these were not fully reported in the published paper. Also, baseline measurements at the beginning of the open-label phase were not reported as a comparison to the end of study (year 7) results indicating reporting bias within the study.

Impact on guideline

The population, intervention and outcomes measured were all of relevance to the clinical guideline. As the reported outcomes do not fully answer those posed by the research recommendation (i.e. bone toxicity and cost effectiveness of monitoring) there is no anticipated impact on the guideline at this time.

This page was last updated: 19 October 2017