Limitations of serum ferritin to predict liver iron concentration responses to deferasirox therapy in patients with transfusion-dependent thalassaemia

Background: In transfusion-dependent anaemias, while absolute serum ferritin levels broadly correlate with liver iron concentration (LIC), relationships between trends in these variables are unclear. These relationships are important because serum ferritin changes are often used to adjust or switch chelation regimens when liver magnetic resonance imaging (MRI) is unavailable. Objectives and methods: This post hoc analysis of the EPIC study compared serum ferritin and LIC in 317 patients with transfusion-dependent thalassaemia before and after 1 yr of deferasirox. Results: Serum ferritin responses (decreases) occurred in 73% of patients, 80% of whom also have decreased LIC. However, 52% of patients without a serum ferritin response did decrease LIC and by >1 mg Fe/g dw (median 3.9) in 77% of cases. Absolute serum ferritin and LIC values correlated significantly only when serum ferritin was <4000 ng/mL (r = 0.59; P < 0.0001) and not at higher levels (≥4000 ng/mL; r = 0.19). Serum ferritin response was accompanied by decreased LIC in 89% and 70% of cases when serum ferritin was <4000 or ≥4000 ng/mL, respectively. Conclusions: As serum ferritin non−response was associated with LIC decrease in over half of patients, use of liver MRI may be particularly useful for differentiating true from apparent non-responders to deferasirox based on serum ferritin trends alone.