Safety and efficacy of stopping tenofovir disoproxil fumarate in patients with chronic hepatitis B following at least 8 years of therapy: a prespecified follow-up analysis of two randomised trials.

Summary Background Effective and well tolerated nucleos(t)ide analogue treatment exists for patients with chronic hepatitis B, although treatment is generally anticipated to be life-long, with concomitant costs and treatment-related side-effects. We aimed to characterise the outcomes of patients with persistent viral suppression who discontinued nucleotide analogue use after extended treatment. Methods The primary objective of this prespecified analysis was to evaluate the safety of stopping long-term tenofovir disoproxil fumarate therapy in patients enrolled in two (completed) randomised controlled studies, GS-US-174-0102 (ClinicalTrials.gov, number NCT00117676) and GS-US-174-0103 (ClinicalTrials.gov, number NCT00116805). In those studies, patients who had completed 8 years or more of nucleotide analogue treatment, were hepatitis B surface antigen (HBsAg)-positive with hepatitis B virus (HBV) DNA concentration of less than 29 IU/mL, and were unwilling or unable to continue therapy were required by protocol to enter a 24-week treatment-free follow-up (TFFU) phase. We present data for patients in the TFFU phase who were assessed at baseline and monitored every 4 weeks for changes in qualitative serum HBsAg, HBV DNA, and alanine aminotransferase (ALT) concentrations in addition to standard safety assessments. Findings Of 124 patients who entered the TFFU phase, 54 (44%) patients did not complete 24 weeks of follow-up (median 12 weeks; IQR 0–20). Overall, 32 (26%) patients reported an adverse event. Serious adverse events occurred in five (4%) patients, including elevated ALT concentrations in two patients, hepatic flare in two patients, and increased lipase in one patient. 38 (31%) of patients had grade 3 or higher laboratory abnormalities, the majority of which were ALT elevations (36 patients). Of the 106 hepatitis B e antigen (HBeAg)-negative patients who entered the TFFU phase, 63 (59%) were followed for 24 weeks. HBsAg loss was observed in five (5%) of the 106 HBeAg-negative patients who entered the TFFU phase, and 37 (35%) had both HBV DNA concentrations of less than 2000 IU/mL and ALT concentrations less than the ULN at TFFU week 24. 18 HBeAg-positive patients entered the TFFU phase, of whom seven (39%) were followed up for 24 weeks. Of these seven patients, none had HBsAg loss or HBV DNA of less than 2000 IU/mL and one (14%) had an ALT less than the ULN at week 24. Interpretation Within 24 weeks of stopping 8 years or more of nucleotide analogue therapy almost a third of patients experienced a grade 3 or higher laboratory abnormality. Although few patients achieved HBsAg loss, a subgroup of HBeAg-negative patients can achieve a low-replicative state within a short duration of follow-up. Funding Gilead Sciences, Inc.