Introduction The kinetics of serum HBsAg (qHBsAg) decline predict response to treatment with pegylated interferon (Peg-IFN) in chronic hepatitis B (CH-B). Whereas nucleos(t)ide analogue (NA) therapy is associated with a sharp decrease in serum HBV DNA level due to inhibition of the viral polymerase, only limited data are available comparing the effect of different treatment regimens on qHBsAg kinetics. This study compared serum qHBsAg and HBV DNA kinetics in patients with CH-B receiving de-novo therapy with either tenofovir (TDF) 245 mg/day, entecavir (ETV) 0.5 mg/day or lamivudine 100 mg/day plus adefovir 10 mg/day (LAM+ADV).
Methods 205 CH-B therapy naïve monoinfected patients (75% males, 21% HBeAg positive, 21% cirrhotic, median age 36 years) were treated with TDF (n=50), ETV (n=62) or LAM+ADV (n=93) for at least 12 months. We quantified qHBsAg (Abbott ARCHITECT® assay) and HBV DNA (real-time TaqMan PCR) in serial serum samples at baseline, months 3 (M3), 6 (M6) and 12 (M12).
Results Median qHBsAg levels were similar at each time-point and were not influenced by treatment regimen. No patient achieved HBsAg loss. More patients receiving TDF achieved optimal virological response (VR), defined as HBV DNA <20 IU/ml), at M3 compared to ETV and LAM+ADV (60% vs 43% vs 40%, respectively; p=0.05) but not at M6 (76% vs 74% vs 79%) and M12 (80% vs 82% vs 82%). The proportion of patients achieving HBsAg decline >0.5 log10 IU/ml was higher in ETV cohort at M3 than in TDF and LAM+ADV (10% vs 2% and 2% respectively; p=0.05), but not at M6 (10% vs 6% vs 5%) and M12 (12% vs 10% vs 8%). HBeAg loss was more frequent in LAM+ADV and TDF groups than ETV (47% vs 33% vs 6%, respectively; p=0.03) and correlated with greater qHBsAg decline at all treatment time-points (M3: r=0.47, p=0.04; M6: r=0.55, p=0.03 & M12: r=0.58, p=0.03). Patients achieving VR had higher qHBsAg levels (M3: 3.71 vs 3.29 log10IU/ml; M6: 3.67 vs 3.24; and M12: 3.66 vs 3.24; all p<0.01), slower qHBsAg decline and fewer patients with qHBsAg >0.5 log10IU/ml from baseline compared to those with detectable HBV DNA.
Conclusion Serum qHBsAg kinetics during therapy were a good predictor of HBeAg loss. However, antiviral therapy in CH-B with nucleos(t)ide analogues in the first 12 months was similar between variable therapeutic approaches and there was no significant qHBsAg decline in contrast to HBV DNA suppression.
Competing interests M Horner: None declared, M Bruce: None declared, S Knighton: None declared, M Al-Freah: None declared, D Joshi: None declared, S Hughes: None declared, A Suddle: None declared, P Harrison: None declared, K Agarwal: None declared, I Carey grant/research support from: BMS, Gilead.