Background and aims <p>Hepatitis B surface antigen (HBsAg) seroclearance is an ideal treatment end point in chronic hepatitis B (CHB), linked to improved clinical outcomes. Baseline HBsAg level is a key predictor, but pooled evidence on the annual incidence of spontaneous HBsAg clearance in untreated CHB patients, stratified by baseline HBsAg levels, is lacking. We conducted a systematic review and meta-analysis to quantify this incidence.</p> Methods <p>Six databases were searched for cohort studies and randomized trials of untreated adult CHB patients reporting baseline HBsAg levels and annual clearance rates. Data were collected and analyzed using the random-effects model.</p> Results <p>Ten studies involving 17,439 patients with 16,134.82 person-years of follow-up were included. The overall pooled annual incidence of spontaneous HBsAg clearance was 1.52% (95% CI 1.11–1.98%), with high heterogeneity (<i>I</i>^<i>2</i> = 94.8%). Excluding three outlier studies reduced heterogeneity (<i>I</i>^<i>2</i> = 47.2%) and yielded a revised incidence of 1.22% (95% CI 1.06–1.39%). Stratification by baseline HBsAg revealed a pronounced gradient: 3.83% (95% CI 2.73–5.10%) for &lt; 100 IU/mL, 1.34% (95% CI 1.14–1.56%) for 100–1000 IU/mL, and 0.51% (95% CI 0.32–0.73%) for &gt; 1000 IU/mL.</p> Conclusions <p>Spontaneous HBsAg clearance remains a relatively uncommon event in untreated CHB patients, yet its incidence is highly dependent on baseline HBsAg levels, with a higher likelihood of clearance at lower levels. These findings provide benchmark data for clinical decision-making regarding antiviral therapy initiation and for the design of future clinical trials. Current evidence is limited, underscoring the need for more high-quality, finely stratified studies in this area.</p>