Moji Awogbade

The linear effects of alpha-thalassaemia, the UGT1A1 and HMOX1 polymorphisms on cholelithiasis in sickle cell disease

Serum bilirubin levels and predisposition to gallstones in sickle cell disease (SCD) are influenced by genetic variation in the hepatic uridine diphosphate (UDP)‐glucuronosyltransferase (UGT1A1) gene, but the association is not consistent. This study investigated whether variation in the gene encoding haem oxygenase (HMOX1), a rate‐limiting enzyme upstream of UGT1A in the haem catabolic pathway, and α‐thalassaemia could explain some of the inconsistent effects. The UGT1A1 [TA]n and HMOX1 [GT]n promoter polymorphisms and α globin genotypes were determined in 263 SCD patients (199 HbSS, 5 HbS/β0, 59 HbSC). Detection of gallstones was based on ultrasound of the liver/biliary tree. Regression analysis showed that serum bilirubin levels and the incidence of gallstones were strongly associated with the number of UGT1A1 [TA] repeats in all subjects (P < 0·0001 and P < 0·01, respectively). While HMOX1 genotype had no effect, co‐inheritance of α‐thalassaemia reduced serum bilirubin levels in all SCD patients independently of the number of UGT1A1 [TA] repeats. Each additional [TA] repeat is associated with an increase in mean serum bilirubin levels of 21% and cholelithiasis risk of 87% in SCD.

Blood transfusion usage among adults with sickle cell disease - a single institution experience over ten years

Transfusion of red blood cells is a major therapeutic option in sickle cell disease (SCD). There is strong evidence for its efficacy, particularly in primary and secondary stroke prevention in children, however, its use in other areas remains controversial. This study assessed the patterns of transfusion in the adult cohort attending King’s College Hospital over a 10‐year period, from 2000 to 2009. Total blood usage has increased significantly (P = 0·006) during this time, with 78% of the blood received by only 6% of the patients. The increase is explained by increased automated red cell exchange and increased usage for planned and acute transfusions for sickle‐related complications.

Guideline on the management of acute chest syndrome in sickle cell disease

1. Department of Haematology, Guy’s and St Thomas’ NHS Foundation Trust 2. Lane Fox Respiratory Unit, St Thomas’ Hospital, Guy’s and St Thomas’ NHS Foundation Trust 3. Department of Haematology, North Middlesex Hospital 4. Department of Paediatric Haematology, Bristol Royal Hospital for Children 5. Department of Haematological Medicine, King’s College Hospital 6. Department of Haematology, Whittington Hospital

Survival in adults with sickle cell disease in a high-income setting

To the editor: Survival of patients with sickle cell disease (SCD) in high-income countries has improved greatly in the last 60 years. In 1960, it was described as a “disease of childhood”[1][1] whereas 25 years later, the Cooperative Study of Sickle Cell Disease reported that 85% of hemoglobin

Association of sickle avascular necrosis with bone morphogenic protein 6

Dear Editor, Avascular necrosis (AVN) of femoral and humeral heads is a frequent and debilitating complication in patients with sickle cell disease (SCD), its prevalence being highest in individuals with SCD-Hb SS and coincidental α-thalassemia. Family and sibling studies suggest a genetic predisposition, and recently, variation in genes involved in bone modeling or the vasculature have been proposed as significant factors in the development of AVN [1]. Baldwin et al. [1] investigated single nucleotide polymorphisms (SNPs) in candidate genes involved in vascular function, inflammation, oxidant stress, and endothelial cell biology for association with AVN. SNPs in bone morphogenic protein 6 (BMP6), annexin A2 (ANXA2), and klotho (KL) genes were suggested to be associated with sickle cell AVN. We attempted to replicate this association in 244 adult patients with SCD, 39 of whom had joint symptoms with radiological and/or magnetic resonance imaging (MRI) evidence of AVN (AVN group; cases), whereas the remaining 205 had no clinical symptoms of AVN (nonAVN group; controls). Those with AVN, on average, were 3 years older than the controls (Table 1). In addition, individuals with AVN had a higher prevalence of coincidental α-thalassemia, higher hematocrit levels, and lower fetal hemoglobin (Hb F) levels when compared to the nonAVN group, though these observations were not statistically significant (Table 1). All patients were of West African or African Caribbean descent. The study was approved by the King’s College Hospital Research Ethics Committee (LREC 01-083). Based on the visual inspection of phased haplotypes in the Yoruba population from Ibadan, Nigeria (YRI), two haplotype blocks were defined in KL, one block in ANXA2, and three blocks in BMP6. Tag SNPs with population frequency information were then selected using the HapMap Project data provided for the Yoruba community, as this ethnic group relates most closely to our sample population. Of the selected tag SNPs, one per gene was identical to that used by Baldwin et al. (Table 2). The selected tag SNPs were genotyped using the TaqMan® allelic discrimination assay. Most assays performed well, but ANXA2-2 failed repeatedly. Genotype data generated from the SNP assays for the remaining six markers ranged from 95.2% to 100% complete, and markers displayed no deviation from the Hardy–Weinberg equilibrium. To test for trait association with the candidate SNPs, allele frequencies between cases (AVN group) and controls (non-AVN group) were compared using Pearson’s chisquare statistic with a significance threshold of p≤0.05 (Table 2). Characterization of the six SNP markers revealed that only one (BMP6-3 or rs3812163) showed significant evidence of association (p=0.021) with AVN. Ann Hematol (2009) 88:803–805 DOI 10.1007/s00277-008-0659-5

Circulating DNA: a potential marker of sickle cell crisis

Free circulating DNA is present in the plasma of healthy subjects, and is elevated in conditions characterized by increased cell death, such as cancer and physical trauma. Analysis of circulating DNA in plasma could provide a useful biomarker in sickle cell disease (SCD) in view of the increased cell turnover through chronic ongoing haemolysis, recurrent vaso‐occlusion and inflammation. Plasma DNA was determined by real‐time quantitative polymerase chain reaction (PCR) amplification of the β‐globin gene (HBB) in 154 patients with SCD [105 haemoglobin (Hb)SS, 46 HbSC and three HbS/β0 thalassaemia] and 53 ethnically matched controls. Blood samples were obtained from all patients in steady state; 21 of the 154 patients were also sampled during admission to hospital for acute pain. Median concentration of circulating plasma DNA in acute pain was more than 10‐fold that in steady state and in controls – 10070 vs. 841 and 10070 vs. 933 genome equivalents/ml respectively (P < 0·0001, in both cases). During steady state, patients had plasma DNA levels similar to controls. Plasma DNA levels in SCD correlated with C‐reactive protein levels (P < 0·005) and total white cell counts (P < 0·05) in steady state. The study shows that plasma DNA concentration may have potential as a biomarker in sickle cell patients.

Outcome of adults with sickle cell disease admitted to critical care – Experience of a single institution in the UK

Sickle cell disease (SCD) patients are perceived to have a high mortality when admitted to the Critical Care Unit (CCU). We performed a retrospective analysis of all adult sickle admissions to CCU at a single centre over an 8‐year period (1 January 2000 to 31 December 2007). Thirty‐eight patients (14 male) were admitted 46 times to CCU; the commonest reasons for admission were acute chest syndrome (14, 30%), multi‐organ failure (8, 17%) and planned post‐elective surgery (7, 15%). CCU mortality for SCD patients was 19·6%, comparable to a CCU‐wide mortality of 17·6% during the study period in the same institution. Re‐admission to CCU was high (16% over the 8‐year period) but did not increase mortality risk.

The measurement of urinary hydroxyurea in sickle cell anaemia

Hydroxyurea is increasingly used in the treatment of sickle cell disease (SCD) although there is little evidence on how best to monitor treatment and compliance. It is also not known why 10–50% patients do not benefit from the drug and whether some of this resistance is because of pharmacokinetic factors. We have developed an assay using mass spectrometry (MS) to measure urinary concentrations of hydroxyurea. We have used this assay to study 12 children and six adults with SCD taking hydroxyurea and found that urinary hydroxyurea was present for at least 12 h following tablet ingestion. Thirty‐five urine samples were analysed that were expected to contain hydroxyurea, based on the reported timing of the last dose and hydroxyurea was detected in 29 (83%) of these. There were also marked differences in urinary hydroxyurea concentrations, suggesting pharmacokinetic variability may explain some of the differences in response to hydroxyurea. Urine samples were also analysed by MS for penicillin metabolites and 43 of the 57 (75%) contained phenoxyacetate, suggesting the ingestion of penicillin within the last 12 h. These assays are potentially useful to study hydroxyurea metabolism further, develop optimal dosing regimes and monitor compliance with treatment.

Effects of co‐existing α‐thalassaemia in sickle cell disease on hydroxycarbamide therapy and circulating nucleic acids

Hydroxycarbamide (HC) is a key treatment option for sickle cell disease (SCD) (Charache et al, 1995; Ferster et al, 2001; Ware & Aygun, 2009; Voskaridou et al, 2010), but not all patients respond clinically, despite compliance (Steinberg et al, 1997). HC response is monitored by increases in fetal haemoglobin, mean corpuscular volume and mean corpuscular haemoglobin. We have previously shown that such increases were muted in SCD patients with co-inherited a thalassaemia (a-SCD) compared to those without (SCD) (Vasavda et al, 2008). Cell-free DNA (cfDNA), a marker of tissue damage, is elevated during sickle acute pain (Vasavda et al, 2007) and reduced with HC therapy (Ulug et al, 2008). It is generally accepted that co-inherited a thalassaemia reduces haemolysis in SCD, subsequent to which the haematocrit is relatively increased with predisposition to vaso-occlusive complications (Ballas, 2001). We hypothesized that an increase in vaso-occlusive events would, in turn, have an effect on cfDNA levels in SCD. Here we describe the effect of co-inherited a thalassaemia on clinical response to HC, and on cfDNA levels. Fifty-two participants were recruited from the specialist haematology clinics at King’s College (n = 38) and Guy’s and St Thomas’ (n = 14) Hospitals, London. All participants from Guy’s and St Thomas’ hospital were treated with HC. The study was approved by the South London REC Office (08/ H0803/185); all participants gave informed, written consent. Inclusion criteria were regular attendance at the sickle cell outpatient clinics, and homozygous sickle cell disease (HbSS) or compound heterozygotes with b thalassaemia (HbSb). Exclusion criteria were: lack of data on a globin genotype, regular blood transfusion and poor compliance (in the treated group). Commencing or terminating HC therapy was a purely clinical decision, and dosing criteria were as previously described (Ulug et al, 2008). For patients treated with HC, clinical data were collected retrospectively for the year prior to starting HC therapy (‘Pre-HC’) and prospectively throughout the study (‘OnHC’). For patients commencing HC during the study, ‘OnHC’ data were collected from 6 months after starting HC therapy. Clinical response was assessed in terms of attendance at accident and emergency, hospital admissions per year, and days spent as an inpatient per year due to sickle-related pain (‘inpatient days’). Blood samples were collected at each ‘steady state’ clinic visit for cfDNA measurement, and plasma cfDNA extracted and quantified as previously described (Vasavda et al, 2007). Clinical data or blood samples were not collected for 3 months after any pro rata blood transfusion. Data was analysed using an unpaired Student’s t-test to compare data between the a-SCD and SCD groups (significance threshold P = 0Æ05). To avoid skewing of data in favour of patients with a greater number of samples and clinical data points, a ‘patient mean’ was calculated and used in further analyses. Clinical response was assessed for those patients treated with HC. Mean ‘on HC’ and ‘pre HC’ laboratory values were obtained, and response calculated as the magnitude change in each parameter; this change represents an intra-patient change in response to HC. An overall mean was then calculated for each group (SCD and a-SCD). For the cfDNA analysis data was not analysed intra-individually: results were stratified according to HC therapy, then further by a thalassaemia status. Laboratory (haematological) data were available for ‘pre-HC’ and ‘on-HC’, as these are routine care tests, whereas cfDNA is a research test, requiring a separate sample. Only a very small number of patients started HC during the study period, so sufficient pre and post intra-individual cfDNA values were not available for analysis. However, we did show in a previous study that HC therapy resulted in a significant change in cfDNA levels (Ulug et al, 2008). Patients with heterozygous (aa/-a) and homozygous (-a/-a) a thalassaemia were analysed as one group. Mean cfDNA levels in the a-SCD and SCD groups were compared separately for HC-treated and untreated patients. 33 patients were on HC therapy with a mean HC dose of 1151 mg (18Æ3 mg/kg/d); 19 were untreated during the study duration. Participants were followed for a mean of 10Æ4 months (range 3Æ6–19Æ2 months). Three participants required blood transfusions during the study (all in the ‘NoHC’ group). Further patient characterization is shown in Table I. Pre-HC data was not available for all participants; numbers used are shown with each result. HC therapy dramatically reduced the total number of hospital admissions and accident and emergency department (A&E) attendances per year in both groups (SCD and a-SCD) (Table I) but the difference between the groups was not significant. However, a-SCD patients had a significantly (P = 0Æ02) smaller reduction in the number of inpatient days (3Æ833 d ± 3Æ146 N = 7) compared to the SCD group (19Æ08 ± 3Æ932. N = 12) (Fig 1A); changes in haematological indices reflected our previous findings (Table I). correspondence

Hydroxyurea therapy lowers circulating DNA levels in sickle cell anemia

Hydroxyurea reduces the frequency of acute pain in sickle cell disease (SCD). We sought to determine if hydroxyurea therapy affects cell free DNA (cfDNA) levels in SCD. cfDNA levels fell in all 10 patients studied; before hydroxyurea, mean was 1,879 (95% CI 1,104–3,199) GE/mL; after hydroxyurea, mean was 780 (95% CI, 634–959) GE/mL (P = 0.002). Mean cfDNA level in the 10 HbSS adults prior to starting hydroxyurea was also significantly higher than that in 115 HbSS case controls who had never taken hydroxyurea (1,879 vs 975 GE/mL, P = 0.02). cfDNA levels may be useful in monitoring response to hydroxyurea therapy in SCD. Am. J. Hematol., 2008.