Hepatitis B surface antigen (HBsAg) loss is a rare event during nucleos(t)ide analogue (Nuc) therapy. Limited data suggest that stopping Nuc therapy may increase HBsAg loss rate in hepatitis B e antigen–negative patients. A large study was conducted to investigate this issue in more detail. Of the 1,075 hepatitis B e antigen–negative patients treated with Nuc for a median of 156 (61‐430) weeks, 5 showed HBsAg seroclearance during treatment at an estimated annual incidence of 0.15%. Of the patients who remained HBsAg‐seropositive, 691 (52.3 years old, 86% male, 44.6% cirrhosis) had stopped Nuc therapy by the Asian‐Pacific Association for the Study of the Liver stopping rule and then were prospectively followed up. Baseline and on‐treatment clinical and viral features, treatment duration, consolidation duration, time to undetectable hepatitis B virus DNA, time to normal alanine aminotransferase, end‐of‐treatment HBsAg, and HBsAg log reduction were compared between patients with and without HBsAg seroclearance after end of treatment. During a median off‐therapy follow‐up period of 155 (2‐614) weeks, HBsAg seroclearance was confirmed in 42 patients. The 6‐year cumulative incidence was 13% with an estimated annual incidence of 1.78%. Cox regression analysis showed that shorter time to undetectable hepatitis B virus DNA (<12 weeks), greater HBsAg reduction during therapy (>1 log10), lower end‐of‐treatment HBsAg level (<100 IU/mL), patients with sustained response, and relapsers not retreated were factors for off‐therapy HBsAg seroclearance. Conclusion: The incidence of HBsAg seroclearance after stopping Nuc was much higher than that during therapy and highest in patients without virologic and clinical relapse; patients with clinical relapse who remained untreated had a 7.34 times higher incidence of HBsAg clearance than those who received retreatment, suggesting that transient untreated clinical relapse may drive sufficient immune control to functional cure. (Hepatology 2017).