Iron overload across the spectrum of non-transfusion-dependent thalassaemias: role of erythropoiesis, splenectomy and transfusions

Abstract

Non-transfusion-dependent thalassaemias (NTDT) encompass a spectrum of anaemias rarely requiring blood transfusions. Increased iron absorption, driven by hepcidin suppression secondary to erythron expansion, initially causes intrahepatic iron overload. We examined iron metabolism biomarkers in 166 NTDT patients with β thalassaemia intermedia (n = 95), haemoglobin (Hb) E/β thalassaemia (n = 49) and Hb H syndromes (n = 22). Liver iron concentration (LIC), serum ferritin (SF), transferrin saturation (TfSat) and non-transferrin-bound iron (NTBI) were elevated and correlated across diagnostic subgroups. NTBI correlated with soluble transferrin receptor (sTfR), labile plasma iron (LPI) and nucleated red blood cells (NRBCs), with elevations generally confined to previously transfused patients. Splenectomised patients had higher NTBI, TfSat, NRBCs and SF relative to LIC, than non-splenectomised patients. LPI elevations were confined to patients with saturated transferrin. Erythron expansion biomarkers (sTfR, growth differentiation factor-15, NRBCs) correlated with each other and with iron overload biomarkers, particularly in Hb H patients. Plasma hepcidin was similar across subgroups, increased with >20 prior transfusions, and correlated inversely with TfSat, NTBI, LPI and NRBCs. Hepcidin/SF ratios were low, consistent with hepcidin suppression relative to iron overload. Increased NTBI and, by implication, risk of extra-hepatic iron distribution are more likely in previously transfused, splenectomised and iron-overloaded NTDT patients with TfSat >70%.

Keywords: anaemia; ineffective erythropoiesis; iron overload; non-transfusion-dependent thalassaemia.

Figure 1

Correlations between sTfR and (A) Hb (B) NTBI (C) GDF‐15 and (D) NRBCs. Evaluable biomarkers of erythroid expansion generally correlated with each other. The strong negative correlation of Hb levels with sTfR supports the concept that sTfR levels reflect erythroid mass, which increases with greater anaemia. GDF‐15, growth differentiation factor‐15; Hb, haemoglobin; NRBC, nucleated red blood cells; NTBI, non‐transferin‐bound iron; sTfR, soluble transferrin receptor.

Figure 2

Correlations between TfSat and markers of iron storage: (A) LIC and (B) serum ferritin; and markers of iron turnover: (C) NTBI and (D) LPI. The relationship between TfSat and NTBI was continuous, whereas the relationship between TfSat and LPI appeared binary. β TI, β thalassaemia intermedia; dw, dry weight; Hb, haemoglobin; LIC, liver iron concentration; LPI, labile plasma iron; NTBI, non‐transferrin‐bound iron; TfSat, transferrin saturation.

Figure 3

Correlations between NTBI and markers of iron storage: (A) LIC and (B) serum ferritin. Strong correlations were noted between NTBI as a marker of iron turnover and LIC and serum ferritin, markers of iron storage. dw, dry weight; LIC, liver iron concentration; NTBI, non‐transferrin‐bound iron.

Figure 4

Correlations between hepcidin and markers of iron turnover: (A) TfSat and (B) NTBI. Hepcidin correlated with markers of iron turnover, but not markers of iron storage. β TI, β thalassaemia intermedia; dw, dry weight; NTBI, non‐transferrin‐bound iron; TfSat, transferrin saturation.

Figure 5

Correlation between serum ferritin and LIC in splenectomised and non‐splenectomised patients. The relationship between LIC and serum ferritin differed in splenectomised patients, as the serum ferritin increment was greater in relation to LIC. dw, dry weight; LIC, liver iron concentration.

Figure 6

Median ± IQR baseline levels of (A) NTBI, (B) TfSat, (C) GDF‐15 and (D) NRBCs in splenectomised and non‐splenectomised patients. Splenectomised patients generally had higher levels of NTBI, TfSat, GDF‐15 and NRBCs. GDF‐15, growth differentiation factor 15; IQR, interquartile range; NRBC, nucleated red blood cell; NTBI, non‐transferrin‐bound iron; TfSat, transferrin saturation.