Abstract
Background: Hydroxyurea (HU) is a key therapy for sickle cell disease (SCD) that aims to reduce the frequency of vaso-occlusive crises (VOC). Previous studies have established a link between SCD and an increased risk of hematological malignancies; however, research on the risk of solid tumors remains limited. Most existing literature focuses on single-center studies or case series; no large-scale, multicenter, or population-based studies are currently available. We aimed to examine the impact of HU on the incidence of hematological and solid malignancies using a comprehensive, nationally representative database.
Methods: We conducted a retrospective cohort study using the TriNetX research network. Patients with SCD treated with HU within three years of diagnosis were compared to those not (controls). Propensity score matching was used to balance cohorts based on demographics, Charlson comorbidities, baseline hemoglobin levels, and other SCD-modifying therapies (e.g., crizanlizumab, voxelotor, L-glutamine). The primary outcome was the incidence of hematologic and solid malignancies. Secondary outcomes included all-cause mortality, hospitalizations, emergency department (ED) visits, and rates of VOC. Subgroup analyses based on age categories (e.g., 0–18, 18–30, 30–50, and > 50 years) and transfusion status were conducted to account for potential confounding by disease severity. Time-to-event outcomes were analyzed using Kaplan–Meier and Cox proportional hazards models.
Results: 89,348 SCD patients were identified, among whom 15,884 were treated with HU, while 76,547 were not. PSM resulted in 13,927 matched pairs. Patients treated with HU had significantly higher rates of hematologic malignancies (HR 2.16, 95% CI 1.40–3.34) and solid tumors (HR 2.29, 95% CI 1.77–2.96) compared to controls. Subgroup analyses revealed that HU-treated patients aged 18–30 had the highest risks of hematological malignancies (HR 4.01, 95% CI 1.15–13.98) and solid tumors (HR 4.21, 95% CI 1.86–9.49); furthermore, those with transfusion history had a 3.7-fold increased risk in hematological malignancies (HR 3.70, 95% CI 1.55–9.13) and a 2-fold increase in solid tumors (HR 2.11, 95% CI 1.33–3.34). The most common types of malignancies within subgroups were myelodysplastic syndrome (MDS) (0.2%) and skin cancers (0.6%). Patients with SCD using HU had higher rates of mortality (1.8% vs. 2.6%, HR 1.46), hospitalizations (9.6% vs. 21.2%, HR 2.79), ED visits (22.4% vs. 26.4%, HR 1.38), and VOC (17.8% vs. 37.4%, HR 3.37).
Conclusion: This comprehensive retrospective study found that the use of HU is associated with an increased risk of both hematologic and solid cancers in SCD, as well as higher mortality rates and increased healthcare utilization. This association remained significant among non-transfusion-dependent patients, suggesting that disease severity alone may not fully explain these findings. MDS was the most common type of hematological malignancy. Skin cancer was the most common solid tumor found in individuals with SCD treated with HU. These associations raise important safety and benefit considerations concerning HU use. Additional prospective studies including treatment duration, disease genotype, and long-term follow-up are necessary.
