BackgroundData on the incidence of hepatitis B virus reactivation (HBVr) remain limited. In Japan, patients who are HBsAg-negative and anti-HBc-positive with undetectable HBV DNA are monitored without antiviral prophylaxis, allowing the estimation of HBVr incidence.MethodsUsing administrative claims data from Japan (2008-2022), we included patients receiving immunosuppressive or chemotherapeutic agents who were HBV DNA-negative at baseline, underwent at least two HBV DNA tests, and did not receive antiviral therapy. HBVr incidence was estimated and stratified according to AGA guideline-based risk categories.ResultsWe analysed 9422 patients. In the high-risk category, HBVr incidence was 52.3 per 1000 person-years (PY) (95% CI: 37.7-70.7) after anti-CD20 therapy and 46.5 per 1000 PY (95% CI: 15.1-108.5) after other high-risk agents. In the moderate-risk category, incidence was 21.0 per 1000 PY (95% CI: 8.4-43.2) after anthracycline derivatives, 12.1 per 1000 PY (95% CI: 4.9-25.0) after tyrosine kinase inhibitors, and 21.0 per 1000 PY (95% CI: 11.5-35.2) after moderate- to high-dose corticosteroids for ≥ 4 weeks. Low-dose or short-term corticosteroids were also associated with HBVr (12.4 per 1000 PY; 95% CI: 9.0-16.6). HBVr also occurred at 12.5 per 1000 PY (95% CI: 2.6-36.6) following other uncertain-risk immunosuppressive agents, including calcineurin inhibitors.ConclusionsHBVr incidence was quantified across a broad range of immunosuppressive and chemotherapeutic therapies, including those with previously limited incidence data. The observed incidence patterns were broadly consistent with current guideline-based risk stratification, while also highlighting measurable risk in some therapies traditionally considered lower risk.