Monitoring and risk of hepatitis B reactivation and hepatitis flare during tenofovir interruption among people with HIV and hepatitis B

OBJECTIVE: To assess hepatitis B virus (HBV) monitoring, HBV reactivation and hepatitis flares during tenofovir interruptions among people with HIV and HBV. DESIGN: Cohort study of electronic health records. METHODS: All tenofovir (tenofovir disoproxil fumarate and tenofovir alafenamide) interruptions among people with HIV and positive HBV surface antigen (HBsAg) or positive HBV core antibody (HBcAb) were categorized by reactivation risk [high: HBsAg+; moderate: HBsAg-/HBcAb+/surface antibody (HBsAb) negative; low: HBsAg-/HBcAb+/HBsAb+]. Incidence rates of HBV reactivation and hepatitis flare were assessed with Poisson regression. RESULTS: Among 5343 individuals with HIV and HBV, there were 6252 tenofovir interruptions (11% high risk, 19% moderate risk, 69% low risk). During the interruptions, HBV DNA/HBsAg testing was infrequent (high: 52%/25%; moderate: 8%/31%, low: 5%/28%), although alanine transaminase (ALT) testing was performed during nearly all interruptions. The HBV reactivation rate was 9.59 per 100 person-years [95% confidence interval (CI): 7.91-11.64] during high-risk, 0.58 (0.36, 0.91) during moderate-risk, and 0.04 (0.02, 0.11) during low-risk interruptions. The HBV reactivation with hepatitis flare incidence rate was much lower, especially in the high-risk group (3.06 per 100 person-years; 95% CI: 2.19-4.29). CONCLUSIONS: In this large US cohort of people with HIV and HBV, tenofovir interruptions were common and HBV monitoring was sub-optimal. HBV reactivation rates were highest among the high-risk group, but much lower among the moderate-risk and low-risk groups. However, some reactivations were likely missed due to low monitoring frequency. Primary and HIV care providers must incorporate HBV monitoring in their standard of care and proceed with caution if considering a tenofovir interruption for people with HIV and HBV.