Level of Hepatitis B (HB) Core Antibody Associates With Seroclearance of HBV DNA and HB Surface Antigen in HB e Antigen-Seronegative Patients.

ABSTRACT

BACKGROUND & AIMS: Although a low level of hepatitis B surface antigen (HBsAg) is a marker of hepatitis B virus (HBV) seroclearance, additional biomarkers are needed for more accurate prediction. We investigated whether quantification of antibody against HBV core protein (anti-HBc) can identify patients with undetectable levels of HBV DNA and HBsAg seroclearance among those who were HBV e antigen (HBeAg)-seronegative. METHODS: We performed a retrospective analysis of data from a community-based cohort of individuals (30-65 years old) in Taiwan who were HBsAg seropositive, anti-HCV negative, and free of cirrhosis and/or liver cancer, recruited from 1991 through 1992, and evaluated every 6-12 months until June 30, 2004. We measured levels of anti-HBc in blood samples from 2500 participants who were seronegative for HBeAg. The first date at which a sample tested negative for HBV DNA or HBsAg, and remained negative in subsequent tests, was designated as the date of spontaneous HBV DNA undetectability or HBsAg seroclearance. We calculated cumulative incidences of HBV DNA undetectability and HBsAg seroclearance; associations between level of anti-HBc and undetectability of HBV DNA or HBsAg seroclearance were estimated by Cox proportional hazard regression. The effects of time on the associations between level of anti-HBc and HBsAg seroclearance was assessed by the area under the receiver operating characteristic curve (AUROC) analysis. RESULTS: After a 12-year follow-up period, higher proportions of subjects with levels of anti-HBc <3 log IU/mL had undetectable levels of HBV DNA (58%) and HBsAg seroclearance (53%) than subjects with higher levels of anti-HBc (29.6% and 19.8%, respectively) (P < .001). For subjects with levels of HBsAg <102 IU/mL and anti-HBc <3 log IU/mL, the adjusted rate ratio of HBV DNA undetectability was 16.45 (95% CI, 11.15-24.28) and of HBsAg seroclearance was 17.95 (95% CI, 12.49-25.81), compared to subjects with higher levels of HBsAg and anti-HBc. A model that included level of anti-HBc as a parameter identified subjects with HBsAg seroclearance within 10 years with an AUROC of 82%; this value was significantly higher than that from models that include only level of HBV DNA and HBsAg (P < .0001). CONCLUSIONS: In a retrospective analysis of a large cohort of patients with chronic HBV infection in Taiwan, we associated levels of anti-HBc <3 log IU/mL with undetectable HBV DNA and HBsAg seroclearance occurred within 10 years; patients who also have levels of HBsAg <102 IU/mL have greater odds. Combining data on levels of HBsAg, HBV DNA, and anti-HBc is able to identify HBeAg-seronegative patients who can achieve HBsAg seroclearance with an AUROC value of 82%.