Hepatitis B surface antigen (HBsAg) decrease and serum interferon-inducible protein-10 levels as predictive markers for HBsAg loss during treatment with nucleoside/nucleotide analogues.

BACKGROUND Hepatitis B surface antigen (HBsAg) loss is the ultimate goal of antiviral therapy and its prediction may be important for treatment individualization. Quantitative HBsAg (qHBsAg) has been shown to predict response to interferon-α, but few studies have analysed qHBsAg during treatment with nucleoside/nucleotide analogues (NAs). Serum interferon-inducible protein-10 (IP-10) has been associated with treatment response in hepatitis C, but data in chronic hepatitis B are lacking. Here, we aimed to investigate potential factors predictive for HBsAg loss. METHODS HBsAg was quantified at multiple time points in 126 patients with chronic HBV infection; 95 received NA treatment for 6-107 months. At an early time point (first 6 months of therapy) and late time point after virological response (VR; HBV DNA<100 IU/ml), we distinguished three patterns of HBsAg decrease: strong decrease (>0.5 log(10)), moderate decrease (10% to 0.5 log(10)) and no decrease (<10%). In addition to conventional biochemical and virological parameters, we analysed serum IP-10 levels in 55 patients. RESULTS Early and late HBsAg kinetics did not correlate. Overall, 42% of patients with a strong HBsAg decrease 2 years after VR cleared HBsAg. Importantly, no patient without a late HBsAg decrease >0.5 log(10) cleared HBsAg. By contrast, early HBsAg decrease after 6 months of NA therapy was not associated with HBsAg loss. Baseline serum IP-10 levels were associated with late but not early HBsAg kinetics and were highest in patients with HBsAg loss. CONCLUSIONS Monitoring qHBsAg after successful HBV DNA suppression might be useful to identify patients who clear HBsAg, implicating finite NA treatment. The role of IP-10 as predictive marker for HBsAg loss should be further evaluated.