Capabilities of hepatitis B surface antigen are divergent from hepatitis B virus DNA in delimiting natural history phases of chronic hepatitis B virus infection

Abstract

ObjectiveQuantitative hepatitis B surface antigen (HBsAg) and hepatitis B virus (HBV) DNA in the natural history of chronic HBV infection have not been rationally evaluated. This study aimed to re-characterize quantitative HBsAg and HBV DNA in the natural history phases.MethodsA total of 595 and 651 hepatitis B e antigen (HBeAg)-positive patients and 485 and 705 HBeAg-negative patients were assigned to the early and late cohorts, respectively. Based on the ‘S-shape’ receiver operating characteristic (ROC) curves, the HBeAg-positive sub-cohorts with possibly high HBV replication (PHVR) and possibly low HBV replication (PLVR) and the HBeAg-negative sub-cohorts with possibly high HBsAg expression (PHSE) and possibly low HBsAg expression (PLSE) were designated.ResultsThe areas under the ROC curve (AUCs) of HBsAg and HBV DNA in predicting HBeAg-positive significant hepatitis activity (SHA) in the early cohort, sub-cohort with PHVR, and sub-cohort with PLVR were 0.655 and 0.541, 0.720 and 0.606, and 0.553 and 0.725, respectively; those in the late cohort, sub-cohort with PHVR, and sub-cohort with PLVR were 0.646 and 0.501, 0.798 and 0.622, and 0.603 and 0.674, respectively. The AUCs of HBsAg and HBV DNA in predicting HBeAg-negative SHA in the early cohort, sub-cohort with PHSE, and sub-cohort with PLSE were 0.508 and 0.745, 0.573 and 0.780, and 0.577 and 0.729, respectively; those in the late cohort, sub-cohort with PHSE, and sub-cohort with PLSE were 0.503 and 0.761, 0.560 and 0.814, and 0.544 and 0.722, respectively. The sensitivity and specificity of HBsAg ≤4.602 log10 IU/ml in predicting HBeAg-positive SHA in the early cohort were 82.6% and 45.8%, respectively; those in the late cohort were 87.0% and 44.1%, respectively. The sensitivity and specificity of HBV DNA >3.301 log10 IU/ml in predicting HBeAg-negative SHA in the early cohort were 73.4% and 60.8%, respectively; those in the late cohort were 73.6% and 64.1%, respectively.ConclusionQuantitative HBsAg and HBV DNA are valuable, but their capabilities are divergent in delimiting the natural history phases.

A critical review of diagnostic and prognostic markers of chronic hepatitis B infection

“…For HBV e antigen-positive patients only, there is no significant association found between qHBsAg and HBV DNA levels [ 17 ]. One plausible therapeutic objective in the management of CHB is HBsAg seroclearance, which is linked to lower qHBsAg levels [ 18 ]. Nevertheless, because HBsAg seroclearance occurs at a low yearly rate, attempts are being made to increase it and use biomarkers like qHBsAg to forecast it early.…”

“…Furthermore, the illness stage may affect the efficacy of HBsAg as a diagnostic sign. A study indicated that the late cohort had a poorer sensitivity of HBsAg in predicting severe hepatitis activity than the early cohort [ 18 ]. As a result, although HBsAg is a useful marker for detecting the hepatitis B virus, its limitations should be taken into account.…”

“…In HBeAg-negative individuals, HBsAg levels at the end of therapy can also predict a persistent off-treatment response; lower levels are linked to a decreased likelihood of virological resurgence [ 59 ]. HBV DNA levels and quantitative HBsAg have different capacities for defining the natural history stages of long-term HBV infection [ 18 ].…”