The role of hepatitis B surface antigen quantification in predicting HBsAg loss and HBV relapse after discontinuation of lamivudine treatment

Abstract

We investigated whether the quantification of hepatitis surface antigen (HBsAg) could predict HBsAg loss or hepatitis B virus (HBV) relapse after stopping lamivudine treatment. A total of 188 naive chronic hepatitis B patients (83 HBeAg-positive, 105 HBeAg-negative patients), who were previously treated with lamivudine (treatment duration: 89.3 ± 35.9 weeks, range: 52-243 weeks) but stopped the treatment for at least 12 months were recruited. The cumulative incidence of HBsAg loss and HBV relapse at year 6 after stopping lamivudine treatment was 24% and 65.9% respectively. Cox regression analysis revealed that lower alanine aminotransferase (ALT) at baseline, lower HBsAg levels at the end of treatment, and longer treatment duration were independent predictors for HBsAg loss, and old age, male sex and higher HBsAg levels at the end of treatment were independent predictors for post-treatment HBV relapse. At the end of treatment, the HBsAg cut-off value of 300 IU/ml could predict 55.6% (5/9) HBsAg loss in HBeAg-positive patients. In HBeAg-negative patients, the HBsAg cut-off values of 120 and 200 IU/ml could predict 79.2% (19/24) HBsAg loss and 93.3% (28/30) post-treatment sustained response respectively. Further HBsAg reduction (>0.22 log IU/ml) at month 6 after stopping treatment was an independent predictor for HBsAg loss after adjusting for HBsAg level at the end of treatment. Serum HBsAg level at the end of treatment is a useful predictor to guide the timing of stopping lamivudine treatment in chronic hepatitis B patients.