Incidence of clozapine-associated neutropenia in patients with schizophrenia: 20-year population-based study in Hong Kong

Background Clozapine is the gold standard for treatment-resistant schizophrenia, but its use is constrained by the risk of severe neutropenia, thus requiring mandatory lifelong hematological monitoring in many jurisdictions. However, a significant evidence gap persists regarding the long-term comparative risk profiles of clozapine versus other second-generation antipsychotics (SGAs), complicating shared decision-making. Aims To compare the risk of neutropenia associated with clozapine versus non-clozapine SGAs, assess temporal risk evolution and identify high-risk subgroups to optimise monitoring. Method This retrospective cohort study used Hong Kong’s electronic health records to identify patients initiating clozapine or non-clozapine SGAs (2003–2019), with follow-up until 2023. Outcomes included minor (absolute neutrophil count (ANC) 1.0–1.5 × 10 9 /litre) and serious neutropenia (ANC <1.0 × 10 9 /litre). Serious neutropenia leading to clozapine discontinuation within 6 weeks was considered clozapine-associated. Results Among 4868 clozapine and 38 277 non-clozapine SGA users, clozapine users showed higher unadjusted incidences of minor (7.05 v . 3.74%) and serious neutropenia (2.53 v . 1.45%). However, adjusted rates of serious neutropenia were similar between groups (0.37 v . 0.36 per 100 person-years). Clozapine was associated with higher risks of minor (incidence rate ratio (IRR) 5.91; 95% CI 3.96–8.70) and serious neutropenia (IRR 3.48; 1.70–6.84) in the first 18 weeks. Bayesian change-point analysis showed sharply declining clozapine-associated risk after 26–29 weeks, converging with non-clozapine SGAs by 82–135 weeks (around 2 years). No significant early excess risk was found in patients younger than 45, while males experienced a transient risk of minor neutropenia. Conclusions The risk of clozapine-associated neutropenia is highest in the initial 18 weeks, declining to levels seen with non-clozapine SGAs by 2 years. Weekly monitoring for 18 weeks and monthly for up to 2 years is advisable, with potential for personalised schedules based on age and sex.