Abstract
The vast majority of patients with myelodysplastic syndrome (MDS) present with anemia, and many of them become transfusion dependent in the long term. Although transfusion iron is primarily taken up by the reticuloendothelial cells, the metal is later redistributed to parenchymal cells, and a portion of MDS patients develop parenchymal iron overload, which may have a negative impact on survival (
N Engl J Med 2005;352:536–8
). The redistribution of transfusion iron from macrophages to parenchymal cells is likely to be modulated by hepcidin levels and erythroid activity. In fact, hepcidin prevents the release of iron from macrophages, and its synthesis is partly down-regulated by erythroid activity, which varies considerably within MDS patients. Expanded but ineffective erythropoiesis is the major mechanism responsible for anemia in low-risk MDS patients, particularly in those with refractory anemia with ringed sideroblasts, while erythroid marrow hypoproliferation is generally found in high-risk patients, typically in those with excess of blasts. We studied 76 patients with MDS followed at the Department of Hematology Oncology, University of Pavia & Fondazione IRCCS Policlinico San Matteo, Pavia, Italy. The WHO classification criteria were employed for diagnosis of MDS. Median time since diagnosis was 12 months (range 0–241). Twenty-five patients were RBC transfusion-dependent (median number of RBC units received was 15, range 4–160). Erythroid activity was evaluated through measurement of soluble transferrin receptor (sTfR) and serum Epo, while body iron status was assessed through serum iron, TIBC and serum ferritin (sFtn). Serum hepcidin was quantified by laser desorption ionization time-of-flight mass spectrometry preceded by weak cation-exchange chromatography exploiting des-Asp hepcidin (hepcidin-24) as an internal standard (www.hepcidinanalysis.com or PLoS ONE 2008;16;3:e2706
). sTfR levels were found to be independently associated with hemoglobin (Hb) (the higher Hb, the lower sTfR; P<.001), serum Epo (the higher Epo, the lower sTfR; P<.001), and WHO category (patients with purely erythroid disorders having higher values than those with multilineage dysplasia and excess blasts; P<.001), as well as with sFtn (the higher sFtn, the lower sTfR; P=.02). A wide variability in hepcidin level was found in MDS patients (median 6.98 nM, range 0.18–92.05 nM). Patients with pure erythroid disorders had significantly lower hepcidin levels compared with those with multilineage dysplasia or excess of blasts (median values 4.41, 7.62, and 15.31 nM, respectively; P<.001). Transfusion-dependent patients had significantly higher hepcidin levels compared with transfusion-independent subjects (15.31 vs 4.73 nM, P<.001). Significant linear correlations were found between serum hepcidin and Hb (r=−.28, P=0.02), serum Epo (r=.44, P<.001), sFtn (r=.68, P<.001) and sTfR (r=−.64, P<.001). Considering all MDS patients, multivariable analysis showed that serum hepcidin levels were independently determined by sTfR levels (the higher sTfR, the lower serum hepcidin; P<.001) and sFtn (the higher sFtn, the higher serum hepcidin; P<.001). Restricting multivariable analysis to MDS patients receiving regular blood transfusion, the presence of ring sideroblasts and their number were also found to have an independent, negative impact of serum hepcidin levels (P=.005). These findings suggest that in MDS patients the redistribution of transfusion iron from reticuloendothelial cells to parenchymal cells is influenced by erythroid activity through its effect on serum hepcidin levels. MDS patients with expanded but ineffective erythropoiesis have low levels of hepcidin and enhanced iron release from macrophages, and therefore a higher likelihood of parenchymal iron loading. In particular, patients with refractory anemia with ring sideroblasts show low hepcidin levels in spite of iron overload, and therefore appear to be at high risk of parenchymal organ damage. Since these this condition has a benign clinical course, preventing organ damage through iron chelation therapy appears clinically important.Disclosures: No relevant conflicts of interest to declare.
Author notes
Corresponding author
2008, The American Society of Hematology
2008
