Genetic Predictors of Hydroxyurea Response in Children with Sickle Cell Disease.

Hydroxyurea therapy has proven laboratory and clinical efficacy for ameliorating the signs and symptoms of sickle cell anemia, primarily by increasing the level of fetal hemoglobin (HbF). Virtually all treated pediatric patients will have symptomatic benefit and improvement in laboratory parameters associated with disease severity, including increased HbF and decreased white blood cell (WBC) count and lactate dehydrogenase (LDH). However, significant inter-patient variability exists in the maximum tolerated dose (MTD) of hydroxyurea and some patients have exaggerated toxicity (excessive myelosuppression) at relatively low doses. Similarly, there is substantial variation in the degree of response for relevant laboratory parameters such as HbF and LDH. We hypothesized that genetic polymorphisms may play an important role in the observed inter-patient variability for both hydroxyurea response and toxicity.

Pediatric patients enrolled in the prospective Hydroxyurea Study of Long-term Effects (HUSTLE, NCT00305175) were eligible for analysis, including 72 subjects who had reached hydroxyurea MTD based on standardized criteria with dose escalation to mild bone marrow suppression. Candidate genes were selected based on a presumed effect on hydroxyurea pharmacokinetics (PK) or pharmacodynamics (PD) such as putative HbF modifiers, ribonucleotide reductase, and urea transporters. For these candidate genes, sequencing was performed for single nucleotide polymorphism (SNP) discovery. In addition, published genetic modifiers were studied to validate the proposed effect on hydroxyurea responses. Each SNP was first tested for Hardy-Weinberg equilibrium and then tested for association with three variables: MTD HbF, MTD LDH, and MTD hydroxyurea dose (mg/kg/day) using Least Squares Regression in dominant, co-dominant, and recessive genetic models.

The average age (mean ± 1SD) at hydroxyurea treatment initiation was 8.4 ± 4.9 years, with 58% males. The average MTD dose was 25.8 ± 4.2 mg/kg/day (median 26.7, range 14.2 – 35.5 mg/kg/day). The average MTD laboratory values for this cohort included hemoglobin (Hb) = 9.7 ± 1.1 gm/dL, mean corpuscular volume (MCV) = 110 ± 12 fL, HbF = 24.5 ± 7.4%, and LDH 432 ± 158 U/L, all significantly different than baseline values (p <0.001). In the process of SNP discovery within the candidate genes, 331 genetic variants were identified including 124 novel SNPs; an additional 14 previously published SNPs in 5 genes were also included. Based on the projected allele frequency, haplotype associations, and type of genetic variant, a total of 70 unique SNPs in 18 genes with allele frequency ≥ 0.10 were included in the final genetic analyses. There were 19 SNPs identified with statistically significant associations for at least one MTD variable (6 SNPs with effect on MTD HbF, 7 with effect on MTD LDH, and 9 with effect on MTD dose), with p-values ranging from 0.0015 to 0.041. Genetic polymorphisms in the UTB, RRM2, BCL11A, and HbS1L-MYB genes had multiple significant associations.

In the prospective HUSTLE study that includes standardized hydroxyurea dose escalation to MTD, significant associations were identified between candidate gene variants and the MTD phenotype for HbF, LDH, and hydroxyurea dose. SNPs in candidate genes previously reported to be important in basal HbF regulation were significant for MTD dose and novel SNPs in the urea transporter UTB were also significantly associated with MTD HbF. These results support the hypothesis that genetic modifiers play an important role in the treatment and toxicity responses to hydroxyurea therapy for children with SCA.

Off Label Use: Hydroxyurea for children with sickle cell anemia.