Background: The serum level of GDF15 has been recently indicated as a possible marker of erythropoiesis (Tanno et al., Nature 2007) suggesting a role of its over-expression in contributing to iron overload in thalassemia syndromes by inhibiting hepcidin expression. The aim of present study has been to evaluate GDF15 serum levels in a homogeneous series of thalassemia patients and the relationship with transfusional parameters and iron status markers.

Methods: A group of consecutive patients with beta thalassemia major followed at our institution were included in the study. All patients were on regular transfusion and iron chelation treatment. Quantification of GDF15 on serum samples was performed with DuoSet ELISA for human GDF15 (R&D Systems) following the manufacturer’s protocol (Tanno et al., Nature 2007). Each patient had also a blood test for haemoglobin (Hb), serum iron, ferritin, transferrin, transferrin saturation and EPO levels. Liver Iron Concentration by SQUID and cardiac iron by MRI T2* have been assessed. The mean hemoglobin levels of the previous year (pre-transfusional, post-transfusional and mean) have been calculated for each individual. The presence of mild thalassemic mutations was used to classify mild or severe genotype. Clinical status has been assessed on the presence/absence of main complications (heart disease, liver disease, diabetes, hypothyroidism). Statistical analysis was performed using the software Statistica (StatSoft).

Results: One hundred-forty patients (73 male, 67 females) were studied. The mean age was 27.9 ± 9.0 years (range: 3.5–42). One hundred (71%) were splenectomised. Betathalassemia major patients had elevated GDF15 serum levels (mean 6892 ± 6894 pg/mL; range 720–52521) in comparison with healthy volunteers (273 ± 104 pg/mL; range 129–401). GDF 15 levels were strongly related to EPO levels (r=0,81; p<0,001). GDF15 levels were not related with age, gender, spleen, clinical status and iron markers. Patients with a severe genotype had higher GDF15 levels than mild genotype patients. GDF15 levels had a negative correlation with Hbs (p<0,05 for actual Hb and pre-transfusional Hb; p<0,001 for post-transfusional Hb and mean Hb). In thalassemia major patients with a severe genotype, GDF15 levels within thrice the normal range have been observed only in patients with pre-transfusional Hb above 9,6, post-transfusional Hb above 12,5 and a mean Hb above 11,3.

Conclusions: In beta thalassemia major patients on regular transfusion and iron chelation, serum GDF15 levels are high, inversely related to the haemoglobin levels maintained. Further studies of this marker may lead to a rethinking of the optimal transfusion therapy in these conditions.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author