The Pharmacogenomics Journal (2007) 7, 386–394 & 2007 Nature Publishing Group All rights reserved 1470-269X/07 $30.00 www.nature.com/tpj ORIGINAL ARTICLE

Fetal in sickle cell anemia: genetic determinants of response to hydroxyurea

QMa1, DF Wyszynski1, The increase in (HbF) in response to hydroxyurea (HU) 1 2 1 varies among patients with sickle cell anemia. Twenty-nine candidate JJ Farrell , A Kutlar , LA Farrer , within loci previously reported to be linked to HbF level (6q22.3–q23.2, 3,1 CT Baldwin and 8q11–q12 and Xp22.2–p22.3), involved in metabolism of HU and related to MH Steinberg1 erythroid progenitor proliferation were studied in 137 sickle cell anemia patients treated with HU. Three-hundred and twenty tagging single 1Department of Medicine, Boston University nucleotide polymorphisms (SNPs) for genotyping were selected based on School of Medicine, Boston, MA, USA; HapMap data. Multiple linear regression and the nonlinear regression 2Department of Medicine, Medical College of Georgia, Augusta, GA, USA and 3Center for Random Forest method were used to investigate the association between Human Genetics, Boston University School of SNPs and the change in HbF level after 2 years of treatment with HU. Both Medicine, Boston, MA, USA methods revealed that SNPs in genes within the 6q22.3–23.2 and 8q11–q12 linkage peaks, and also the ARG2, FLT1, HAO2 and NOS1 genes were Correspondence: Dr MH Steinberg, Center of Excellence in Sickle associated with the HbF response to HU. Polymorphisms in genes regulating Cell Disease, E248, Boston Medical Center, 88 HbF expression, HU metabolism and erythroid progenitor proliferation might E. Newton Street, Boston, MA 02118, USA. modulate the patient response to HU. E-mail: [email protected] The Pharmacogenomics Journal (2007) 7, 386–394; doi:10.1038/sj.tpj.6500433; published online 13 February 2007

Keywords: SNPs; association analysis; sickle cell; fetal hemoglobin; hydroxyurea

Introduction

Fetal hemoglobin (HbF) inhibits the polymerization of sickle hemoglobin (HbS).1 As many of the complications of sickle cell anemia (homozygosity for HBB, glu6val), like osteonecrosis, acute chest syndrome and painful episode, are associated with the level of HbF, and, HbF is inversely associated with mortality, investigators have assiduously sought pharmacological means of increasing HbF production.2–6 Hydroxyurea (HU), a ribonucleotide reductase inhibitor, is one drug that increases HbF concentration in patients with sickle cell anemia7–10 and it is the sole FDA-approved agent for treating sickle cell anemia. Most, but not all patients respond to HU treatment with an increase in HbF, but as with the baseline HbF concentration, which varies widely among patients, the magnitude of the HbF response to HU is also variable.10–13 The regulation of HbF level might be a complex genetic trait governed by genetic elements linked to the b- -like cluster and quantitative trait loci (QTL) present on chromosomes 6, 8 and on the X-chromosome; other regulatory loci are also likely to exist and epigenetic and cellular factors could also have regulatory roles.14–27 It is possible that these and other regulatory elements also modulate the HbF response to HU. Received 6 June 2006; revised 19 September 2006; accepted 6 November 2006; published Accordingly, we hypothesized that single nucleotide polymorphisms (SNPs) online 13 February 2007 in candidate genes or QTL with putative roles in the regulation of HbF SNPs, hydroxyurea and HbF in sickle cell anemia QMaet al 387

production might modulate the HbF response to treatment SNP in a gene showing an association when the P-value for with HU. We therefore studied the association of SNPs in significance was set at 0.05. Tables 2 and 3 present the these loci with response to HU in patients who participated statistically significant results of the quantitative trait in the Multicenter Study of Hydroxyurea (MSH), a trial analysis for single SNP association with the change of HbF designed to evaluate the efficacy of this drug in sickle cell level after a 2-year treatment, expressed as percentage of anemia. total hemoglobin and as the absolute HbF level, expressed as g/dl, respectively. An analysis was also performed expressing the increment in HbF after HU treatment as F-cells. Results Seventeen SNPs were significantly associated with the change in percent HbF (Table 2). They included two in The distribution of HbF for 137 patients enrolled in this MAP3K5, five in TOX, two in NOS1, three in FLT1, two in study is shown as Figure 1. The change of HbF after HU ARG2 and two in NOS2A. The most significant association treatment did not follow a normal distribution and was observed with SNP rs2182008 (P ¼ 0.003) in FLT1 resembled a bimodal distribution with a large portion of (Fms-related tyrosine kinase 1), a vascular endothelial growth people with minor or no change (mean ¼ 0) and a small factor, which is involved in cell proliferation and differentia- portion of people with extreme change (mean 440). This tion. Twenty SNPs were significantly associated with the distribution suggested that categorizing these data, like change of absolute HbF (Table 3); a similar pattern of dividing subjects into quartiles by the HbF change, and association was observed and the most significant association comparing patients in lowest quartile group vs those was in SNP rs10483801 (P ¼ 0.0013) in ARG2 (arginase type II, patients in the highest quartile of change, might be an involved in drug metabolism of HU). Using F-cells as the alternative approach for analyzing these data. However, outcome measure gave similar results (data not shown). considering the relatively small sample size in our study, this For candidate genes with significant association in multi- approach provides very limited power for detecting genetic ple SNPs, haplotype associations were explored using associations. Haplo.stats (version 1.2.1) as given in the R library (available Three-hundred and twenty tagging SNPs in 29 candidate at http://cran.us.r-project.org).28 However, as these SNPs genes (Table 1) were examined in 137 sickle cell anemia studied here are tagging SNPs and most of them are not in patients treated with HU. We considered an SNP to have a linkage disequilibrium (LD) with each other, we did not find significant association with response to HU treatment when improved association by haplotype analysis in any genes the P-value was 0.01, unless there was more than one p (data not shown). The results of joint analysis of all the SNPs and covariates (age, sex and the a- and b-globin gene cluster haplotypes) using Random Forest analysis are shown in Figures 2 and 3. The relative importance of one independent variable (a SNP) is measured by %IncMSE (see Materials and methods for details), and the larger the value of %IncMSE, the higher importance that variable has for correct prediction of HbF response to HU. This analysis revealed that the most important variables for predicting the change of HbF level matched most of the SNPs identified by SNP association analysis. Interestingly, SNPs within ASS (arginino- succinate synthetase) and ARG1 (arginase, liver) were observed to have strong effect on the change of HbF level, which was not detected by single SNP association analysis. This suggests that these two genes might be involved in interaction with other genes to regulate the response to HU treatment. SNP rs2182008 in FLT1 showed a strong effect on response to HU treatment. This SNP was significantly associated with the change in HbF under a dominant model (P-value ¼ 0.003 for HbF in percentage and 0.002 for HbF in g/dl) and it was also a highly ranked predictor for response to HU from the Random Forest analysis (second for HbF in percentage and third for HbF in g/dl). The A allele of this SNP was associated with increased HbF level after HU treatment; there is no difference between AA and AG genotypes and the increase in HbF in subjects with these genotypes was on average 5.9 Figure 1 Distributions of HbF change (a) in percentage (%) and (b)in times higher than that in subjects with the GG genotype grams (g/dl). (Figure 4).

The Pharmacogenomics Journal SNPs, hydroxyurea and HbF in sickle cell anemia QMaet al 388

Table 1 Candidate genes selected

Gene Chromosomes Function Tagging SNPs

Cytochrome P450, family 4, subfamily A, polypeptide 11 (CYP4A11) 1 Drug metabolism 4 Hydroxyacid oxidase 2 (long chain) (HAO2) 1 Drug metabolism 5 Kinase insert domain receptor (a type III receptor tyrosine kinase) (KDR) 4 Cell differentiation 17 Arginase, liver (ARG1) 6 Drug metabolism 3 Phosphodiesterase 7B (PDE7B) 6 Chr6 QTL 17 Microtubule-associated 7 (MAP7) 6 Chr6 QTL, cell differentiation 3 Mitogen-activated protein kinase kinase kinase 5 (MAP3K5) 6 Chr6 QTL 10 Peroxisomal biogenesis factor 7 (PEX7) 6 Chr6 QTL 6 NADPH oxidase 3 (NOX3) 6 Drug metabolism 17 Met proto-oncogene (hepatocyte growth factor receptor) (MET) 7 Cell proliferation 5 Nitric oxide synthase 3 (endothelial cell) (NOS3) 7 NO production 4 Glutathione reductase (GSR) 8 Drug metabolism 6 CCAAT/enhancer binding protein (C/EBP), delta (CEBPD) 8 Regulation of DNA 2 transcription Transcription elongation factor A (SII), 1 (TCEA1) 8 Chr8 QTL, regulation of 3 DNA transcription SRY (sex determining region Y)-box 17 (SOX17) 8 Chr8 QTL, regulation of 3 DNA transcription Thymus high mobility group box protein TOX (TOX) 8 Chr8 QTL, regulation of 57 DNA transcription Argininosuccinate synthetase (ASS) 9 NO production 24 P450, family 2, subfamily C, polypeptide 9 (CYP2C9) 10 Drug metabolism 4 Beta-globin gene cluster 11 Chr11 QTL 8 Nitric oxide synthase 1 (neuronal) (NOS1) 12 NO production 23 Fms-related tyrosine kinase 1 (vascular endothelial growth factor/vascular 13 Cell proliferation and 32 permeability factor receptor) (FLT1) differentiation Arginase, type II (ARG2) 14 NO production 7 Aquaporin 9 (AQP9) 15 Drug metabolism 8 NADPH oxidase, EF-hand calcium binding domain 5 (NOX5) 15 Drug metabolism 6 Nitric oxide synthase 2A (inducible, hepatocytes) (NOS2A) 17 NO production 6 EGF-like-domain, multiple 6 (EGFL6) X Chr X QTL 9 Glycoprotein M6B (GPM6B) X Chr X QTL 21 C-fos induced growth factor (vascular endothelial growth factor D) (FIGF) X Chr X QTL 6 Pirin (iron-binding nuclear protein) (PIR) X Chr X QTL 4

Abbreviations: NO, nitric oxide; QTL, quantitative trait loci; SNPs, single nucleotide polymorphisms.

Discussion similar heterogeneity might account for the varied response in HbF among sickle cell anemia patients treated with HU. HU can increase HbF concentration in most individuals with We found statistically significant associations of the HbF sickle cell anemia but some patients who take the drug response to HU with multiple SNPs in several genes. TOX exactly as directed by experienced physicians either fail to (thymus high-mobility group box protein) is located within respond or have an HbF response that might not be the 8q11–q12 linkage peak that Garner et al.,16,33 found to clinically significant. Even among patients with an increase interact with the À158 CT 5’ Gg-globin gene SNP and that in HbF, the magnitude of this increase varies.10–13 The cause might also effect the g-tob-globin gene switch. A member of this variability is poorly understood.11,12,29,30 of the high-mobility group (HMG) box protein family, HbF concentration in is determined by interactions TOX contains a single HMG box motif and binds DNA in a among chromosome remodeling activities, transcription sequence-specific manner. All HMG box are able to factors, genes modulating erythropoiesis, genetic elements induce a sharp bend in DNA. Multiple SNPs in PDE7B linked to the b-globin gene cluster, the kinetics of erythroid (phosphodiesterase 7B) within 6q22–q23 QTL were also cell differentiation and differential red cell survival (for associated with the HbF response to HU. SNPs in this gene reviews see Bank31 and Stamatoyannopoulos32). This were previously reported to be associated with baseline HbF complex regulatory environment provides ample opportunity level in patients with sickle cell anemia.21 for genetic modulation of HbF production. Because of Nitric oxide (NO) binds and activates sGC, which evidence suggesting that heterogeneity in genetic elements increases cGMP production. cGMP interacts with transcrip- modulate HbF baseline concentration, we reasoned that tion factors increasing the expression of the g-globin

The Pharmacogenomics Journal SNPs, hydroxyurea and HbF in sickle cell anemia QMaet al 389

Table 2 SNPs associated with a significant change in HbF%

Genetic model

SNP Chromosomes Gene a Function Codominant Dominant Recessive rs10494225 1 HAO2 Untranslated 0.015 NS 0.0039 rs9376230 6 MAP3K5 Intron NS NS 0.036 rs9483947 6 MAP3K5 Intron NS NS 0.034 rs826729 8 TOX Intron NS 0.045 NS rs765587 8 TOX Intron NS 0.031 NS rs9693712 8 TOX Intron 0.019 0.0098 NS rs172652 8 TOX Intron 0.049 NS NS rs380620 8 TOX Intron NS 0.047 NS rs816361 12 NOS1 Intron NS NS 0.045 rs7977109 12 NOS1 Intron NS NS 0.029 rs9319428 13 FLT1 Intron NS NS 0.047 rs2182008 13 FLT1 Intron 0.012 0.003 NS rs8002446 13 FLT1 Intron 0.033 NS 0.011 rs10483801 14 ARG2 Intron 0.026 NS 0.0075 rs10483802 14 ARG2 Intron 0.015 NS 0.0038 rs1137933 17 NOS2A Synonymous NS NS 0.031 rs944725 17 NOS2A Intron NS 0.02 NS

Abbreviations: HbF, fetal hemoglobin; NS, nonsignificant; SNPs, single-nucleotide polymorphisms. aSee Table 1 for full names. NS: P-value 40.05. Bold and italics: SNP with the most significant P-value.

Table 3 SNPs associated with a significant change in HbF (g/dl)

Genetic model

SNP Chromosomes Gene a Function Codominant Dominant Recessive rs10494225 1 HAO2 Untranslated 0.0078 NS 0.0018 rs2327669 6 PDE7B Intron NS NS 0.041 rs11154849 6 PDE7B Intron NS 0.05 NS rs9376173 6 PDE7B Intron NS NS 0.049 rs1480642 6 PDE7B Intron NS NS 0.044 rs487278 6 PDE7B Intron NS NS 0.017 rs2693430 8 TOX Intron NS 0.049 NS rs765587 8 TOX Intron NS 0.044 NS rs12155519 8 TOX Intron NS 0.037 NS rs9693712 8 TOX Intron 0.048 0.02 NS rs380620 8 TOX Intron 0.038 0.016 NS rs7309163 12 NOS1 Intron NS NS 0.038 rs7977109 12 NOS1 Intron NS NS 0.023 rs3751395 13 FLT1 Intron NS NS 0.039 rs9319428 13 FLT1 Intron NS NS 0.044 rs2182008 13 FLT1 Intron 0.0085 0.0021 NS rs2387634 13 FLT1 Intron NS NS 0.037 rs8002446 13 FLT1 Intron 0.029 NS 0.01 rs10483801 14 ARG2 Intron 0.0052 NS 0.0013 rs10483802 14 ARG2 Intron 0.014 NS 0.0037

Abbreviations: HbF, fetal hemoglobin; NS, nonsignificant; SNPs, single nucleotide polymorphisms. aSee Table 1 for full names. NS: P-value 40.05. Bold and italics: SNP with the most significant p-value. genes.34 ASS (argininosuccinate synthetase) is an enzyme synthases. SNPs in ASS gene were strong predictors for HbF that catalyzes the penultimate step of the arginine bio- response to HU. Significant associations were also found synthetic pathway. Arginine is the substrate for the NO with SNPs in ARG2 and ARG1, genes coding for enzymes

The Pharmacogenomics Journal SNPs, hydroxyurea and HbF in sickle cell anemia QMaet al 390

Figure 2 Joint analysis of multiple SNPs with change in HbF in percentage (%): only predictors that have relative importance value (%IncMSE) 43.0 are shown here; *: SNPs in x-QTL. that hydrolyze arginine to ornithine. SNPs in NOS1 nature of our study samples. Nevertheless, these SNPs (12q24.2-q24.31, neuronal NO synthase) and NOS2A,an were associated with HbF concentration, for example, the NO synthase expressed in liver, were also associated with increase of HbF in subjects carrying A allele of SNP HbF response to HU. rs2182008 was almost six times higher than that in subjects FLT1 (vascular endothelial growth factor (VEGF)/vascular with GG genotype. permeability factor receptor) is a receptor for VEGF. SNP To identify genetic factors regulating the expression of rs2182008 in this gene was found to be strongly associated HbF in sickle cell anemia at baseline and in the absence of with the HbF response to HU. This gene has tyrosine protein treatment with HU, we studied a sample of 327 subjects and kinase activity that is important for the control of cell an independent group of 987 subjects. We replicated in part proliferation and differentiation.35 our previous work21 using this expanded sample of patients Most SNPs associated with HbF response to HU treatment and denser SNP coverage, by finding multiple tagging SNPs were in either untranslated portions of the genes or in in genes abutting the 6q22.3–q23.2 QTL. We also found introns. Most likely, the associations we found are in LD a significant association of HbF with SNPs in TOX and with the actual functionally important elements. Although within EGFL6, GPM6B and FIGF in the Xp22.2–p22.3 QTL most associations have biologically plausible mechanisms (unpublished data). by which they might influence the expression of the According to the common disease–common variant g-globin genes or the concentration of HbF, our studies hypothesis, many of the genetic variants causing complex are hypothesis generating rather than mechanistic, and diseases or phenotypes are expected to have only a we were unable to do functional studies because of the small effect on disease outcome.36 The power to obtain

The Pharmacogenomics Journal SNPs, hydroxyurea and HbF in sickle cell anemia QMaet al 391

Figure 3 Joint analysis of multiple SNPs with change in HbF in grams (g/dl): only predictors that have relative importance value (%IncMSE) 43.0 are shown here; *: SNPs in x-QTL.

typical thresholds of P-value significance after applying Materials and methods multiple testing corrections is limited for such markers, because significance is a function of sample size, allele Patients frequency and effect size. Considering our relatively DNA samples and laboratory data from adult African- small sample size, which nevertheless constitutes the Americans with sickle cell anemia who participated in the largest number of patients with sickle cell anemia taking MSH were analyzed. This was a randomized, placebo- HU under controlled conditions, we validated our findings controlled, double-blind trial designed to test whether HU by using a different analytical method but did not could reduce the number of vasoocclusive events in adults apply multiple testing corrections for reporting our results. with moderate to severe sickle cell anemia.37 HU was given Also, as we selected haplotype tagging SNPs that tag daily, in a single dose starting at 15 mg/kg, and increased by different LD blocks and almost all our associations are 5 mg/kg every 12 weeks up to a maximum-tolerated dose with more than a single SNP in a gene, it is unlikely that the of 35 mg/kg, unless toxicity developed. When toxicity result with one SNP merely reflects the same association occurred, treatment was stopped until blood counts recov- signal with other SNPs. Therefore, out of the 320 SNPs ered. HU was then resumed at a dose 2.5 mg/kg lower than tested, it is less likely that multiple hits with SNPs in the the toxic dose. The MSH enrolled 299 patients, of which 152 same gene is a chance event. Further replication with fewer were randomly assigned to HU. The minimum length of candidate genes and denser SNP coverage may be a more follow-up evaluation for patients with HbF measured at the practical way to reduce false discovery and confirm our end of the study was 21 months (maximum, 38 months; findings. mean, 28). DNA samples were available and HbF was

The Pharmacogenomics Journal SNPs, hydroxyurea and HbF in sickle cell anemia QMaet al 392

study population have .43–48 SNPs that had more than 25% missing genotypes or less than 5% minor allele frequency were not considered in the analysis. This resulted in 280 SNPs being tested for association.

Statistical analysis Single SNP association was investigated via multiple linear regression analysis (SAS v9.1) with simultaneous adjustment for age, sex and the a- and b-globin gene cluster haplotypes. Three genetic models (codominant, dominant and reces- sive) governing modulation of response to HU treatment were tested. In the codominant system, an allele causes the homozygous form to look different from the wild-type and the heterozygous form (all three look different from each other). In dominant transmission, an allele causes the homozygous and the heterozygous forms to look the same as each other, but different from the wild type. In a recessive model, an allele affects the phenotype if it is present in the homozygous state only. One subject exceeding the top third standard deviation of all the values of the change in HbF levels was omitted from the analysis to prevent the extreme values from biasing the results. Figure 4 HbF change for SNP rs2182008 (a) in percentage (%) and (b) We used the nonlinear regression Random Forest method in grams (g/dl). median; D: mean. to identify additional predictors of response to HU treat- ment and to validate our results from single SNP associa- tion.49 Random Forest consists of a collection of regression measured in 137 of the 152 patients randomized to receive trees, each regression tree itself being a regression function. HU. Although pill counts and measurements of serum HU Each of these trees predicts a real value by querying a set levels were used to assess compliance with treatment, the number of variables and instances within the regression rapid clearance of HU from the circulation prevented us model. Each regression tree is thus trained on a different from knowing with certainty if a patient took all of the bootstrap sample of both training instances and features. medication prescribed. The studies reported here were The Random Forest then averages the predictions made by approved by the IRB of Boston Medical Center. the trees in the forest to produce the final output. Random Forest is a variance reduction technique and has provable Laboratory studies properties with regard to resisting over fitting. Additionally, HbF was measured by alkali denaturation at least twice Random Forest is very efficient to train and test, and has pretreatment and twice post-treatment.10 The laboratory built-in mechanisms for estimating test error and con- methods used to define the hemoglobin phenotype and the fidence in each prediction made. This procedure is non- haplotypes of the b-like and a -globin gene cluster were parametric, not model based, and identifies those described previously.11,38,39 independent variables that best segregate subgroups as important predictors and identifies interactions among SNP selection and genotyping independent variables. The relative importance of the The selection of genes was based on previously reported independent variables (e.g. SNPs in our case) is defined as linkage peaks (6q22–q23, 8q11–q12 and Xp22.2–p22. follows. The mean of squared residuals (MSE) is computed 3)16,19,40 genes involved in metabolism of HU and genes on the out-of-bag data for each tree, and then the same is related to erythroid progenitor proliferation. Thirty-three computed after permuting a variable. The differences are candidate genes were chosen in our study, and 320 tagging averaged over all trees and normalized by the standard error, SNPs were selected based on genotype data from the and this calculation is repeated for each independent HapMap project (Phase I, Yoruba sample)41 using Haploview variable. A large difference in MSE (%IncMSE) indicates that (v3.2)42 (Table 1). Genotyping was carried out using a the independent variable is important for correct prediction, custom 384 multiplex design using an Illumina platform. whereas a small difference indicates that the independent For quality control purposes, about 3% of the DNA samples variable is less important for correct prediction. Relative were re-genotyped, and Hardy–Weinberg equilibrium was importance provides a measure by which predictors can be assessed for each SNP. Hardy–Weinberg equilibrium was ranked with respect to each other. Our analyses were per- determined before analysis and was performed for quality formed using the Random Forest (v 4.5–16) package as given control purposes rather than to evaluate if the genotypes in the R library (available at http://cran.us.r-project.org), met Mendelian expectation because all members of our except that, as the number of trees to be grown in the

The Pharmacogenomics Journal SNPs, hydroxyurea and HbF in sickle cell anemia QMaet al 393

forest was set at 2500, we used default parameters of this 16 Garner CP, Tatu T, Best S, Creary L, Thein SL. Evidence of genetic program. interaction between the beta-globin complex and chromosome 8q in the expression of fetal hemoglobin. Am J Hum Genet 2002; 70: 793–799. Acknowledgments 17 Thein SL, Sampietro M, Rohde K, Rochette J, Neatherall DJ, Lathrop GH et al. Detection of a major gene for heterocellular hereditary persistence We thank the investigators of the MSH who obtained blood samples of fetal hemoglobin after accounting for genetic modifiers. Am J Hum for DNA-based studies and analyzed data from these studies for the Genet 1994; 54: 214–228. study publications cited in the text of this paper. This study was 18 Chang YC, Smith KD, Moore RD, Serjeant GR, Dover GJ. An analysis of fetal hemoglobin variation in sickle cell disease: the relative contribu- supported by NHLBI Grant HL R01 70735 (MHS). tions of the X-linked factor, b-globin haplotypes, b-globin gene number, gender, and age. Blood 1995; 85: 1111–1117. 19 Chang YP, Maier-Redelsperger M, Smith KD, Contu L, Ducroco R, de Duality of Interest Montalembert M et al. The relative importance of the X-linked FCP locus and b-globin haplotypes in determining haemoglobin F levels: a The study sponsor had no involvement in study design, data study of SS patients homozygous for bS haplotypes. Br J Haematol 1997; collection, analysis or interpretation, writing of the paper or the 96: 806–814. decision to submit the paper for publication. 20 Craig JE, Rochette J, Sampietro M, Wilkie AO, Barnetson R, Hatton CS et al. Genetic heterogeneity in heterocellular hereditary persistence of fetal hemoglobin. Blood 1997; 90: 428–434. References 21 Wyszynski DF, Baldwin CT, Cleves MA, Amirault Y, Nolan VG, Farrell JJ et al. Polymorphisms near a chromosome 6q QTL area are associated 1 Poillon WN, Kim BC, Rodgers GP, Noguchi CT, Schechter AN. Sparing with modulation of fetal hemoglobin levels in sickle cell anemia. Cell effect of hemoglobin F and on the polymerization of Mol Biol (Noisy -le-grand) 2004; 50: 23–33. hemoglobin S at physiologic ligand saturations. Proc Natl Acad Sci USA 22 Blobel GA, Weiss MJ. Nuclear factors that regulate erythropoiesis, In: 1993; 90: 5039–5043. Steinberg MH, Forget BG, Higgs DR, Nagel RL, (eds). Disorders of 2 Milner PF, Kraus AP, Sebes JI, Sleeper LA, Dukes KA, Embury SH et al. Hemoglobin Genetics, Pathophysiology, and Clinical Management,1st Sickle cell disease as a cause of osteonecrosis of the femoral head. N Engl edn. Cambridge University Press: Cambridge, 2001, pp. 72–94. JMed1991; 325: 1476–1481. 23 Forrester WC, Takegawa S, Papayannopoulou T, Stamatoyannopoulos 3 Castro O, Brambilla DJ, Thorington B, Reindorf CA, Scott RB, Gillette P G, Groudine M. Evidence for a locus activation region: the formation et al. The acute chest syndrome in sickle cell disease: incidence and risk of developmentally stable hypersensitive sites in globin-expressing factors. The cooperative study of sickle cell disease. Blood 1994; 84: hybrids. Nucleic Acids Res 1987; 15: 10159–10177. 643–649. 24 Grosveld F, Antoniou M, Berry M, de Boer E, Dillon N, Ellis J et al. 4 Platt OS, Thorington BD, Brambilla DJ, Milner PF, Rosse WF, Vichinsky E Regulation of human globin gene switching. Cold Spring Harb Symp et al. Pain in sickle cell disease-rates and risk factors. N Engl J Med 1991; Quant Biol 1993; 58:7–13. 325: 11–16. 25 Tanimoto K, Engel JD. In vivo modulation of human beta-globin gene 5 Platt OS, Brambilla DJ, Rosse WF, Milner PF, Castro O, Steinberg MH switching. Trends Cardiovasc Med 2000; 10: 15–19. et al. Mortality in sickle cell disease. Life expectancy and risk factors for 26 Tuan D, Solomon W, Li Q, London IM. The "beta-like-globin" gene early death. N Engl J Med 1994; 330: 1639–1644. domain in human erythroid cells. Proc Natl Acad Sci USA 1985; 82: 6 Steinberg MH. Therapies to increase fetal hemoglobin in sickle cell 6384–6388. disease. Curr Hematol Rep 2003; 2: 95–101. 27 Weiss MJ, Orkin SH. GATA transcription factors: key regulators of 7 Platt OS, Orkin SH, Dover G, Beardsley GP, Miller B, Nathan DG. hematopoiesis. Exp Hematol 1995; 23: 99–107. Hydroxyurea enhances fetal hemoglobin production in sickle cell 28 Schaid DJ, Rowland CM, Tines DE, Jacobson RM, Poland GA. Score tests anemia. J Clin Invest 1984; 74: 652–656. for association between traits and haplotypes when linkage phase is 8 Dover GJ, Humphries RK, Moore JG, Ley TJ, Young NS, Charache S et al. ambiguous. Am J Hum Genet 2002; 70: 425–434. Hydroxyurea induction of hemoglobin F production in sickle cell 29 Maier-Redelsperger M, de Montalembert M, Flahault A, Neonato MG, disease: relationship between cytotoxicity and F cell production. Blood Ducrocq R, Masson MP et al. Fetal hemoglobin and F-cell responses to 1986; 67: 735–738. long-term hydroxyurea treatment in young sickle cell patients. The 9 Charache S, Dover GJ, Moore RD, Eckert S, Ballas SK, Koshy M et al. French Study Group on Sickle Cell Disease. Blood 1998; 91: 4472–4479. Hydroxyurea: effects on hemoglobin F production in patients with sickle 30 Ware RE, Eggleston B, Redding-Lallinger R, Wang WC, Smith-Whitley K, cell anemia. Blood 1992; 79: 2555–2565. Daeschner C et al. Predictors of fetal hemoglobin response in children 10 Charache S, Terrin ML, Moore RD, Dover GJ, Barton FB, Eckert SV et al. with sickle cell anemia receiving hydroxyurea therapy. Blood 2002; 99: Effect of hydroxyurea on the frequency of painful crises in sickle cell 10–14. anemia. N Engl J Med 1995; 332: 1317–1322. 31 Bank A. Regulation of human fetal hemoglobin: new players, new 11 Steinberg MH, Lu ZH, Barton FB, Terrin ML, Charache S, Dover GJ. Fetal complexities. Blood 2006; 107: 435–443. hemoglobin in sickle cell anemia: determinants of response to 32 Stamatoyannopoulos G. Control of globin hydroxyurea. Multicenter study of hydroxyurea. Blood 1997; 89: during development and erythroid differentiation. Exp Hematol 2005; 1078–1088. 33: 259–271. 12 Bakanay SM, Dainer E, Clair B, Adekile A, Daitch L, Wells L et al. 33 Garner C, Silver N, Best S, Menzel S, Martin C, Spector TD et al. Mortality in sickle cell patients on hydroxyurea therapy. Blood 2005; Quantitative trait locus on chromosome 8q influences the switch from 105: 545–547. fetal to adult hemoglobin. Blood 2004; 104: 2184–2186. 13 Zimmerman SA, Schultz WH, Davis JS, Pickens CV, Mortier NA, Howard 34 Ikuta T, Ausenda S, Cappellini MD. Mechanism for fetal globin gene TA et al. Sustained long-term hematologic efficacy of hydroxyurea at expression: role of the soluble guanylate cyclase-cGMP-dependent maximum tolerated dose in children with sickle cell disease. Blood 2004; protein kinase pathway. Proc Natl Acad Sci USA 2001; 98: 1847–1852. 103: 2039–2045. 35 Wang D, Donner DB, Warren RS. Homeostatic modulation of cell 14 Labie D, Pagnier J, Lapoumeroulie C, Rouabhi F, Dunda-Belkhodja O, surface KDR and FLT1 expression and expression of the vascular Chardin P et al. Common haplotype dependency of high Gg-globin endothelial cell growth factor (VEGF) receptor mRNAs by VEGF. J Biol gene expression and high HbF levels in b- and sickle cell Chem 2000; 275: 15905–15911. anemia patients. Proc Natl Acad Sci USA 1985; 82: 2111–2114. 36 Reich DA, Lander ES. On the allelic spectrum of human disease. Trends 15 Steinberg MH, Voskaridou E, Kutlar A, Loukopoulos D, Koshy M, Ballas Genet 2001; 17: 502–510. SK et al. Concordant fetal hemoglobin response to hydroxyurea in 37 Charache S, Terrin ML, Moore RD, Dover GJ, McMahon RP, Barton FB siblings with sickle cell disease. Am J Hematol 2003; 72: 121–126. et al. Design of the multicenter study of hydroxyurea in sickle cell

The Pharmacogenomics Journal SNPs, hydroxyurea and HbF in sickle cell anemia QMaet al 394

anemia. Investigators of the multicenter study of hydroxyurea. Control 44 Nielsen DM, Ehm MG, Weir BS. Detecting marker-disease association Clin Trials 1995; 16: 432–446. by testing for Hardy–Weinberg disequilibrium at a marker locus. Am J 38 Dozy AM, Kan YW, Embury SH, Mentzer WC, Wang WC, Lubin B et al. Hum Genet 1998; 63: 1531–1540. Alpha-globin gene organisation in blacks precludes the severe form of 45 Deng HW, Chen WM, Recker RR. QTL fine mapping by measuring and alpha-thalassaemia. Nature 1979; 280: 605–607. testing for Hardy–Weinberg and linkage disequilibrium at a series of 39 Steinberg MH, Rosenstock W, Coleman MB, Adams JG, Platica O, linked marker loci in extreme samples of populations. Am J Hum Genet Cedeno M et al. Effects of thalassemia and microcytosis on the 2000; 66: 1027–1045. hematologic and vasoocclusive severity of sickle cell anemia. Blood 46 Goring HH, Terwilliger JD. Linkage analysis in the presence of errors IV: 1984; 63: 1353–1360. joint pseudomarker analysis of linkage and/or linkage disequilibrium 40 Game L, Close J, Stephens P, Mitchell J, Best S, Rochette J et al. An on a mixture of pedigrees and singletons when the mode of integrated map of human 6q22.3-q24 including a 3-Mb high-resolution inheritance cannot be accurately specified. Am J Hum Genet 2000; 66: BAC/PAC contig encompassing a QTL for fetal hemoglobin. Genomics 1310–1327. 2000; 64: 64–76. 47 Terwilliger JD. Inflated false-positive rates in Hardy–Weinberg and 41 The International HapMap Consortium. The International HapMap linkage-equilibrium tests are due to sampling on the basis of rare Project. Nature 2003; 426: 789–796. familial phenotypes in finite populations. Am J Hum Genet 2000; 67: 42 Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization 258–259. of LD and haplotype maps. Bioinformatics 2005; 21: 263–265. 48 Weinberg CR, Morris RW. Testing for Hardy–Weinberg disequilibrium 43 Xu J, Turner A, Little J, Bleecker ER, Meyers DA. Positive results in using a genome single-nucleotide polymorphism scan based on cases association studies are associated with departure from Hardy–Weinberg only. Am J Epidemiol 2003; 158: 401–403. equilibrium: hint for genotyping error? Hum Genet 2002; 111:573–574. 49 Breiman L. Random forests. Machine Learning 2001; 45: 5–32.

The Pharmacogenomics Journal