Thrombotic outcomes in sickle cell disease patients treated with crizanlizumab with or without hydroxyurea: A real-world retrospective cohort analysis using the trinetx database

Abstract Introduction: Crizanlizumab, a humanized monoclonal antibody targeting P-selectin, was approved by the FDA in 2019 to reduce the frequency of vaso-occlusive crises (VOCs) in sickle cell disease (SCD). By blocking P-selectin–mediated adhesion of sickled erythrocytes and leukocytes to the endothelium it helps prevent microvascular occlusion. While its efficacy in reducing VOCs has been demonstrated in clinical trials, its effect on thrombotic risk remains unclear. In this real-world analysis, we assessed the association between crizanlizumab use and thrombotic events in patients with HbSS genotype SCD using the TriNetX database. Methods: We conducted a retrospective cohort study using TriNetX, a federated global research network aggregating de-identified electronic medical records from large healthcare organizations. Adult patients with HbSS genotype and ≥1 documented VOC in the year prior to treatment initiation were included. Patients were stratified into two groups: those who received crizanlizumab and those who did not. Propensity score matching (1:1) was performed to balance cohorts by age, sex, HbS and HbF levels, hydroxyurea use, and comorbidities. The primary outcome was the incidence of thrombotic events at 3-, 6-, 12-, and 36-month post-treatment. Relative risks (RRs) and 95% confidence intervals (CIs) were calculated to compare event rates between groups at each interval. Results: A total of 1,816 patients met the inclusion criteria, with 908 in each matched cohort. The mean age was 29.7 years in the crizanlizumab group and 29.6 years in the control group, with a similar female predominance (61.3% vs. 60.8%). Mean HbS and HbF levels were 59.6% and 8.9% in the crizanlizumab group, compared to 63.0% and 9.5% in the control group. Hydroxyurea use was reported in 74.6% of patients in both cohorts, either alone or in combination with crizanlizumab. Crizanlizumab-treated patients had a significantly higher risk of thrombotic events at all time points: 3 months (RR 2.32, 95% CI 1.74–3.11), 6 months (RR 2.20, 95% CI 1.72–2.81), 1 year (RR 2.28, 95% CI 1.85–2.81), and 3 years (RR 2.15, 95% CI 1.82–2.55), all with p < 0.001. Conclusion In this real-world cohort of patients with HbSS sickle cell disease, crizanlizumab use was associated with a significantly increased risk of thrombotic events, even after adjusting for hydroxyurea use and other confounders. While clinical trials have supported its role in reducing VOCs, these findings raise important questions about its safety profile in broader clinical practice. Careful patient selection and close monitoring may be warranted, and further prospective studies are needed to clarify the thrombotic risks associated with P-selectin inhibition.