Varying 10‐year off‐treatment responses to nucleos(t)ide analogues in patients with chronic hepatitis B according to their pretreatment hepatitis B e antigen status

Abstract

To evaluate the long-term durability and efficacy of nucleos(t)ide analogues (NAs) and to determine the related factors for virological relapse in chronic hepatitis B (CHB) patients. CHB patients who fulfilled the criteria for discontinuing NAs therapy in accordance with the published guidelines were included in the study from December 2001. Virological relapse was defined as serum hepatitis virus B (HBV) DNA >104 copies/mL twice at least 2 weeks apart. A total of 223 CHB patients were enrolled at the time their NAs therapy was discontinued. The 10-year cumulative relapse rate (CRR) in hepatitis B e antigen (HBeAg)-positive patients was statistically lower than that in HBeAg-negative patients (30.9% vs 62.3%, P < 0.001). In the HBeAg-positive group, Cox regression analysis showed that age at cessation (hazard ratio [HR] 1.067, P < 0.001), consolidation therapy (HR 0.958, P = 0.021), and time to HBeAg seroconversion (HR 0.943, P = 0.019) were predictors for relapse. In the HBeAg-negative group, age at cessation (HR 1.040, P = 0.004) and time to HBV DNA negativity (HR 1.246, P = 0.010) were potential predictors for virological relapse. The off-treatment responses to NAs differ in CHB patients with different pretreatment HBeAg status. NA withdrawal is generally safe and feasible in young patients with CHB. Long consolidation periods should be preferred in HBeAg-positive patients to achieve better durability. Benefits of cessation of NAs do not last long in HBeAg-negative CHB patients.