Ten-year follow-up of hepatitis B relapse after cessation of lamivudine or telbivudine treatment in chronic hepatitis B patients.

The high rate of relapse after cessation of nucleos(t)ide analogues (NUCs) treatment in chronic hepatitis B (CHB) patients leads us to reassess the feasibility for off-therapy, but long-term follow-up data are scarce. We assessed the feasibility for off-therapy by a long-term observation of relapse in response to lamivudine (LAM) and telbivudine (LdT). Eighty-six NUC-naive CHB patients, treated with LAM (n = 46) or LdT (n = 40) who reached the guidelines recommended for off-therapy, were followed for up to 10 years. Hepatitis B virus (HBV), viral serology and biochemistries were periodically determined. COX model was used to predict the risk of relapse. A total of 52.3% of patients experienced relapse within a median of 115 months (range, 61-122 months). A total of 93.3% of relapses occurred within 48 months. Relapse rates in hepatitis B e antigen (HBeAg)-positive (n = 56) and HBeAg-negative (n = 30) patients were 39.3% and 76.7%, respectively (p < 0.01). HBeAg-positive patients who achieved an early viral response (EVR), defined as undetectable HBV DNA within 6 months, had a lower relapse rate compared to non-EVR patients (21.4% vs. 59.2%, p < 0.01). EVR patients who had both lower HBV DNA (<10(6) copies/mL) at baseline and lower hepatitis B surface antigen (HBsAg) at end of treatment had a relapse rate of 10.7%. The high relapse rates in CHB patients over this 10-year follow-up make LAM or LdT off therapy infeasible in most of the cases, except in the case of HBsAg loss and/or seroconversion. HBeAg-positive patients with EVR, lower HBV DNA and HBsAg had lower relapse rates and could be good candidates for off-therapy. Long-term monitoring, especially during the first 4 years, is critical for patients off-therapy.