Zihao Xu, Yifan Liu, Liangbin Cheng, Jun Xu
Nucleos(t)ide analogue (NA) discontinuation in chronic hepatitis B (CHB) can produce two clinically divergent trajectories: a minority of patients achieve sustained hepatitis B surface antigen (HBsAg) loss or off-treatment viral control, whereas others develop hepatitis flares without virological benefit. Recent longitudinal studies have begun to define the host immune correlates of these outcomes. This Mini Review separates direct evidence from NA-withdrawal cohorts from mechanistic evidence obtained in other CHB settings. Direct human data suggest that beneficial outcomes are associated with recovery of HBsAg-specific memory B-cell activity, plasmablast expansion, stronger T follicular helper collaboration, and preserved hepatitis B virus (HBV)-specific T-cell function. In contrast, high inhibitory-receptor expression on global T cells, persistent inflammatory chemokine signatures, and greater post-flare alanine aminotransferase variability are associated with non-beneficial flares. immune complexes formed by HBsAg and antibodies to HBsAg provide an additional humoral readout that may reflect both flare risk and the quality of HBsAg decline. These observations are promising but remain exploratory: cohorts are small, flare definitions are heterogeneous, assays are not standardized, and most signatures lack external validation. Immune markers should therefore be viewed as candidate correlates to be integrated with viral biomarkers, liver-disease severity, and close post-withdrawal monitoring rather than as stand-alone criteria for stopping therapy.