Portal de Periódicos da CAPES

Urinary hepcidin excretion in patients with myelodysplastic syndrome and myelofibrosis

2008; Wiley; Volume: 142; Issue: 4 Linguagem: Inglês

10.1111/j.1365-2141.2008.07225.x

1365-2141

Asher Winder, Rafi Lefkowitz, Hussam Ghoti, Merav Leiba, Tomas Ganz, Elizabeta Nemeth, Eliezer A. Rachmilewitz,

Erythropoietin and Anemia Treatment

Many patients with myelodysplastic syndrome (MDS) develop severe anaemia that requires multiple blood transfusions with consequent iron overload. The absorption of iron from the intestine, iron release from hepatic stores and the recycling of iron by macrophages is negatively regulated by hepcidin, the principal iron-regulatory hormone (Nicolas et al, 2002; Ganz & Nemeth, 2006). Hepcidin regulates iron absorption and distribution by binding to ferroportin, the iron exporter of these cells, and inducing its internalization and degradation, (Nemeth et al, 2004). Hepcidin is increased in response to increased body iron levels, transferrin saturation (Ganz & Nemeth, 2006) or inflammation, and is decreased in response to hypoxia, (Nicolas et al, 2002) erythropoietic activity (Adamsky et al, 2004) and oxidative stress (Choi et al, 2007). Patients with congenital haemolytic anaemia, such as thalassaemia, develop severe iron overload both due to multiple blood transfusions and to increased iron absorption. Hepcidin levels are inappropriately decreased in thalassaemia intermedia, whereas in thalassaemia major, hepcidin levels were found to be elevated, presumably due to transfusions which decrease erythropoietic drive (Origa et al, 2007). The present study investigated whether urinary hepcidin levels are also low in patients with various forms of MDS, similar to the findings in thalassaemia intermedia. Approval was obtained from the University of California, Los Angeles (UCLA) and Wolfson Medical Centre institutional review boards for the study. Urine samples were collected during visits to the outpatient department of the Wolfson Medical Centre. The patients lacked any clinical signs of inflammation. Urine samples were obtained from four patients with refractory anaemia (RA), four with refractory anaemia with ringed sideroblasts (RARS), three with refractory cytopenia with multilineage dysplasia (RCMD), two with refractory anemia with excess of blasts I (RAEB I), four with refractory anaemia with excess of blasts II (RAEB II) and four with idiopathic myelofibrosis (MF), with an age range from 54 to 87 years (Table I). Nine patients had received less than 10 units of blood and the rest between 13 and 147 units. Seven patients had ferritin levels above 1000 μg/l, most of whom had a transferrin saturation >50% (Table I). Urine samples were preserved in 0·01% sodium azide solution, and frozen (−20°C) for shipment to UCLA and urinary hepcidin was measured. Urinary creatinine concentrations were measured at the UCLA Clinical Laboratories. Hepcidin was normalized using urinary creatinine concentrations as nanograms hepcidin/milligram creatinine. In previous testing, intra-assay precision was 6%. Spiking urine with synthetic hepcidin from 0 ng/ml to 1600 ng/ml, followed by carboxymethyl ion exchange matrix (CM) extraction and immunodot assay resulted in the recovery of more than 90% of spiked hepcidin. Day-to-day variations of hepcidin levels in the urine of three healthy volunteers were 50 ± 26, 18 ± 8 and 55 ± 22 ng/mg creatinine in collections of 8, 15, or 21 successive days (Papanikolaou et al, 2005). Because the lower detection limit of the urinary hepcidin assay was 5 ng/mg creatinine, a value of <5 ng/mg creatinine was attributed to samples with lower values. The urinary excretion of hepcidin was undetectable (<5 ng/mg creatinine) in three MDS and two MF patients, while in the rest, with the exception of two patients, the levels were low, between 10 and 50 ng/mg creatinine – which is considered to be the average normal range. While this is a small study, the degree of iron overload was not found to be directly related to the subtype or to the number of transfusions but it was more severe in those with RARS. The urinary hepcidin excretion was low in all patients including those with high ferritin or high transferrin saturation levels, and inversely correlated with serum ferritin (r2 = 0·62 on a log-log plot). The slope of the correlation was 0·05, indicating a significant decrease of hepcidin relative to ferritin (Fig 1). The hepcidin levels were higher in the chelated patients (median 93 ng/mg vs. 20 ng/mg creatinine, P = 0·03, Mann–Whitney test) (Fig 1). Correlations of hepcidin with (A) serum ferritin (Pearson correlation R = 0·72, P = 0·0006), (B) haemoglobin (R = −0·52, P = 0·02), (C) the number of transfused RBC units (R = 0·74, P = 0·0007) and (D) transferrin saturation (R = 0·53, P = 0·03). Regression lines and confidence intervals (95%) and are shown on log-log (A) or linear (B–D) scales. Patients undergoing chelation therapy are indicated by empty squares. In this study, hepcidin was suppressed in most of the patients with MDS and in four with MF, despite iron overload, similar to the findings in thalassaemia intermedia. Sera from patients with β thalassaemia have been reported to downregulate hepcidin expression in the hepatic HepG2 cell line, while sera from patients with haemochromatosis induced an increase in hepcidin expression (Weizer-Stern et al, 2006). Somewhat similar results were obtained using sera from patients with MDS, although they were not uniform as were the findings in thalassaemia (Breda et al, 2007). This could be due to the presence of increased quantities of an erythroid regulator, GDF15, identified in the sera of patients with thalassaemia, which was present in lower amounts in sera from patients with MDS (Tanno et al, 2007). Although the number of transfusions was found to increase hepcidin urinary excretion, hepcidin was still inappropriately low, and inversely correlated with serum ferritin. Future studies correlating the subtype of MDS with transfusion requirement and hepcidin levels will clarify the mechanisms of iron overload in these diseases.