Assessment of genotoxicity associated with hydroxyurea therapy in children with sickle cell anemia

Abstract

Hydroxyurea induces fetal hemoglobin, improves laboratory parameters, and ameliorates clinical complications of sickle cell anemia (SCA), but its long-term efficacy and safety in this patient population remain incompletely defined. Although generally considered non-DNA reactive, an important safety concern is that hydroxyurea may indirectly cause genotoxic damage. To better address this safety issue of hydroxyurea in patients with SCA, we measured the production of micronuclei (MN) in red blood cells (RBCs) as a marker of genotoxicity. Blood samples were collected from children with SCA enrolled in the Hydroxyurea Study of Long-term Effects (ClinicalTrials.gov NCT00305175). Flow cytometry quantified circulating MN-containing erythrocyte sub-populations before and during hydroxyurea exposure. The frequency of micronucleated reticulocytes (MN-CD71(+)) and micronucleated mature erythrocytes (MN-RBC) was then tested for associations with laboratory and clinical data. In cross-sectional analysis of 293 blood samples from 105 children with SCA and a median of 2 years of hydroxyurea therapy, exposure to hydroxyurea was associated with significantly increased frequencies of MN-CD71(+) and MN-RBC compared to baseline. The increases were evident by 3 months of therapy, and did not escalate further with up to 12 years of continuous drug exposure. In prospective longitudinal analysis, substantial inter-individual variation in the effect of hydroxyurea on %MN-CD71(+) was observed that was associated with the expected laboratory effects of hydroxyurea. In conclusion, clinically relevant exposure to hydroxyurea is associated with increased MN production consistent with erythroblast genotoxicity but with substantial inter-patient variability. Associations between increased %MN-CD71(+) and laboratory benefits suggest that hydroxyurea effects on MN production may be related to individual patient sensitivity to hydroxyurea within the bone marrow.

Copyright 2010 Elsevier B.V. All rights reserved.

Figures

Figure 1. Cross-sectional analysis of the effect of hydroxyurea on MN production and clearance

A shows the effect of hydroxyurea on %MN-CD71+, reflecting MN production. B shows the effect of hydroxyurea on %MN-RBC levels (clearance). All patients were selected from the HUSTLE study. This included patients enrolled already on hydroxyurea therapy as well as patients beginning hydroxyurea therapy. The number of patients at each time point are: 0 (n=58), 3 months (n=45), 6 months (n=44), 9 months (n=41), 1 year (n=27), 2 years (n=18), 3 years (n=16), 4 years (n=11), 6 years (n=6), 7 years (n=6), 8 years (n=4) and 12 years (n=7). The mean and standard deviation are given for each time point. The non-SCA level is the frequency of the respective MN containing erythrocyte subpopulation in ethnic matched control subjects with no SCA.

Figure 2. Prospective analysis of the effect of hydroxyurea on MN production and clearance

Patients beginning hydroxyurea therapy had a baseline sample and serial samples collected at 3, 6, 9, 12 and 24 months while undergoing hydroxyurea treatment. After 3 months of hydroxyurea exposure, median %MN-CD71+ and %MN-RBC levels significantly increased compared to baseline values and did not further increase with extended exposure. The median levels are given for each time point.