Abstract
Stopping nucleos(t)ide analogues (NAs) can foster hepatitis B surface antigen (HBsAg) clearance, yet comparative data for tenofovir disoproxil fumarate (TDF) and entecavir (ETV) are scarce. We compared HBsAg loss, virological relapse (VR), clinical relapse (CR), and safety after stopping TDF or ETV. Ninety-eight non-cirrhotic, HBeAg-negative adults with viral suppression on TDF or ETV for ≥ 3 years were prospectively followed for ≤ 4 years after allocation to TDF discontinuation (TDF-D; n = 39), ETV discontinuation (ETV-D; n = 27), or continued therapy (n = 32). Median follow-up was 146 weeks. HBsAg clearance occurred in 7 patients (10.6%): 12.8% with TDF-D, 7.4% with ETV-D, and none with continued therapy. All clearers had end-of-treatment HBsAg < 100 IU/mL, which was the only independent predictor of clearance (OR 0.38, 95% CI 0.18-0.83, p = 0.02). By week 144, cumulative VR was 83% with TDF-D versus 56% with ETV-D (p = 0.002); CR was 50% versus 27% (p = 0.02). Six TDF-D patients had severe ALT flares, two were hospitalized, and one died. NA discontinuation yielded greater HBsAg clearance than continued therapy, mainly within the first year, without a significant difference between TDF and ETV. However, TDF cessation caused earlier and more severe relapses. Careful candidate selection and intensive monitoring are essential.