Context: Benefits of HU treatment in thalassemic patients are controversial. Most studies agree that HU decreases transfusion requirements in patients with TI, whereas results in patients with TM are inconsistent.

Objectives: To assess the proportion of good responders to HU and the factors associated with good response.

Methods: We initiated HU treatment in 54 thalassemic patients (9 TI and 45 TM). Median ages at first transfusion were 57 months (32–252) in TI and 13 months (4–60) in TM patients. Median ages at inclusion were 12.4 years (4.4–54.9) in TI and 9.9 years (2.0–21.0) in TM patients. Median numbers of erythrocyte concentrates (EC) transfused the year before HU were 3 (1–9) in TI and 12 (8–16) in TM patients (10–65). Good response was defined as more than 70% decrease of transfusional needs one year after beginning HU, partial response as a decrease between 40 and 70%, and non response as a decrease of less than 40%. Patients were transfused when their hemoglobin (Hb) level was < 6g/dL; not transfused when Hb was > 7 g/dL; decision to transfuse was based on clinical status for Hb level between 6 and 7 g/dL. Mean HU dose was 17 ± 2 mg/kg/d (13–21). Mean follow-up was 29.3 ± 14 months.

Results: Eight (90%) TI and 20 (44.5%) TM patients had a good response to HU. No TI and 9 (20%) TM patients had a partial response to HU. One (10%) TI and 16 (35,5%) TM patients had no response to HU. Clinical and hematological tolerance were good except for one 55 yr old TI patient who developed in the 3rd year of treatment an acute myeloblastic leukemia. Retrospectively, a malignant transformation of a pre-existent myeloproliferative disorder can not be excluded and the responsibility of HU is questionable. A better response to HU was associated with a later beginning of transfusion (p=0.02), a higher pre-HU Hb (p= 0.0004), codon 6(-A) mutation (p= 0.002), and with TI diagnosis (p=0.03). On the contrary, Gγ-158 C/C genotype (Xmn1−/−) (p= 0.0001) was associated with a worse response. After adjustment, decrease of transfusional needs was associated with the number of CE transfused the year before HU treatment (p= 0.001), Xmn1 polymorphism (p= 0.003) and the age at the beginning of HU (p=0.03) but not with the type of thalassemia. In conclusion, HU may represent an alternative to transfusions for both TI and TM patients in countries with limited blood supplies when beginning early.

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author

Sign in via your Institution