BACKGROUND AND AIMS:: Seroclearance of hepatitis B surface antigen (HBsAg) is one of the most important clinical outcomes for chronic hepatitis B treatment trials. Few studies have explored the incidence and determinants of spontaneous HBsAg seroclearance using a long-term follow-up study. This study aims to examine the natural history and predictors of HBsAg seroclearance.
METHODS:: A total of 3087 individuals with chronic hepatitis B virus infection were enrolled between 1991-1992 in this community-based study. Serum samples collected at baseline and follow-up examinations were tested for HBsAg, hepatitis B e antigen (HBeAg), serum HBV DNA levels and anti-HCV serostatus. Cox proportional hazards models were used to estimate HBsAg seroclearance rate ratios associated with various determinants.
RESULTS:: HBsAg seroclearance occurred in 562 participants during 24,829 person-years of follow-up, giving a 2.26% annual seroclearance rate. HBV DNA levels at baseline and follow-up were the most significant predictor of seroclearance. Higher HBV viral loads conferred lower HBsAg seroclearance rates (p<0.001). A spontaneous decrease in follow-up HBV DNA level (>/=3 log) was significantly associated with HBsAg seroclearance, showing an adjusted odds ratio (95% CI) of 4.17 (2.55-6.82). Among those with seroclearance, 95.8% had undetectable HBV DNA prior to seroclearance. Cumulative incidence of HBsAg seroclearance at 60 and 100 months after serum HBV DNA decreased to undetectable levels was 25.8% and 51.3%, respectively.
CONCLUSIONS:: This study reveals various determinants of HBsAg seroclearance, and suggests that a low viral load is an important factor affecting the natural seroclearance of HBsAg, indicating significant clinical implications for the treatment of chronic HBV.