EFFECTS OF BCL11A SNPS ON INTERINDIVIDUAL VARIABILITY OF FETAL HEMOGLOBIN INDUCTION IN PATIENTS WITH SICKLE CELL DISEASE TREATED WITH HYDROXYUREA AT THE CENTRO DE HEMATOLOGIA E HEMOTERAPIA DE GOVERNADOR VALADARES

Abstract

Sickle cell disease (SCD) is an autosomal recessive monogenic disorder that modifies the adult hemoglobin and results in multiple clinical phenotypes. Increases in fetal hemoglobin (HbF) has clinical benefits by reducing the clinical severity of SCD. Hydroxyurea (HU) is a drug used to treat SCD that induces HbF expression, thereby ameliorating hematological and clinical profile. However, HbF response to HU treatment is highly variable among patients. A systematic review indicated that SNPs rs4671393 (A>G) and rs766432 (A>C) of BCL11A gene may help to explain interindividual variability in HbF induction in response to HU. We examined the changes in hematological profile, focused on HbF levels (%) after HU therapy. Next, we examined whether variation in BCL11A SNPs rs4671393 and rs766432 affects HbF levels in patients with SCD before and after HU therapy. We studied 93 patients with SCD (89.3% with SCA HbSS genotype: 10.7% with HbSC genotype) with a mean age of 17.9 ± 11.5y and 51.61% of the cohort were women. We measured the levels of total hemoglobin, hematocrit, global leukometry, reticulocytes (%) and HbF during 18 months of HU therapy (range, 5-32 mg/kg) at the Hematology and Hemotherapy Center of Governador Valadares, Minas Gerais. Genotypes for BCL11A SNPs were determined by TaqMan® allele discrimination assays, and HbF levels were measured by HPLC. We observed an increase in hemoglobin (7.6-9.4 g/dL, p < 0.0001), hematocrit (23.0-28.0%, p < 0.0001), and especially HbF concentration (5.3-15%, p < 0.0001) during the 18 months, of HU therapy. In addition, reticulocytes and overall leukocyte count showed a decrease over the 18 months (3.65-1.9% and 13300-7200 mm³, respectively). Patients carrying genotypes with the minor alleles for both BCL11A SNPs (AG+AA for rs4671393 and AC+CC for rs766432) did not show differences in HbF concentration before and after HU therapy compared to genotypes with wild-type alleles (GG for rs4671393 and AA for rs766432). However, the mean (±S.D.) for HbF concentration was numerically higher for AC+CC genotypes (16.69 ± 8.86%; n = 33) than the AA genotype (13.62 ± 6.96%; n = 60) for the rs766432, and for patients carrying the GA+AA genotypes (16.55 ± 8.77%; n = 34) than the GG genotype (13.65 ± 7.02%; n = 59), although these differences did not reach significant level (rs766432, p = 0.068; and rs4671393, P = 0.0832). For rs4671393, median for HbF levels were higher after than before HU therapy both in GG genotype (5.0 vs. 13.0%; p < 0.0001) and AG+AA genotypes (5.5% vs. 15.85; p < 0.0001). For rs766432, HbF concentration was higher in carriers with AA genotype (5.0 vs. 12.65%; p < 0.0001) and AC+CC genotypes (5.4 vs. 16.0%; p < 0.0001). Our findings do not support the role of BCL11A SNPs rs4671393 and rs766432 in the increase of HbF concentration in response to HU therapy, which may be explained by the small number of patients with SCD analyzed. However, our findings are in line with a systematic review showing that BCL11A SNPs may partially explain the interindividual variability in the increase in HbF expression upon HU therapy. Replication studies should consider increasing the number of patients with SCD analyzed.