Hepcidin: a new tool in the management of anaemia in patients with chronic kidney disease?

The introduction of erythropoiesis-stimulating agents (ESA), such as recombinant human erythropoietin (EPO) has allowed effective treatment of anaemia in patients with chronic kidney disease (CKD). However, the optimal target level of haemoglobin is debated and many patients are resistant to ESA. Hepcidin is a recently discovered low-molecular-weight protein that plays an important role in iron metabolism. Hepcidin may be relevant in CKD and explain the often-observed disbalance in iron metabolism and the resistance to ESA. As such hepcidin could become an important tool to predict ESA responsiveness and to guide treatment with ESA and intravenous iron. In addition, hepcidin has the potential to become a target of treatment. In this review we summarize the pathophysiology of hepcidin and discuss its potential relevance for patients with CKD. Human hepcidin is produced by hepatocytes as an 84-amino-acid (aa) pre-prohepcidin [ 1 ]. Subsequent posttranslational processing results in the biologically active 25 aa form (hepcidin-25) that is secreted in the plasma. Additional amino-terminal degradation results in the production of two smaller isoforms (hepcidin-20 and -22), the biological significance of which is unknown [ 2 ]. Hepcidin is the key regulator of systemic iron homeostasis. Hepcidin leads to internalization and degradation of the iron exporter ferroportin, which is present on the cell surface of macrophages and enterocytes [ 3,4 ]. Thus, hepcidin inhibits the release of iron by macrophages and attenuates the iron uptake in the gut. In addition to these effects on body iron distribution, hepcidin might also directly inhibit erythroid-progenitor proliferation and survival [ 5 ]. Increased iron stores and inflammation induce hepcidin production, whereas hypoxia, anaemia, iron deficiency, increased erythropoiesis and ESA attenuate hepcidin synthesis [ 6–10 , Figure 1]. Thus, inflammation decreases the availability of iron, whereas hypoxia or anaemia increases iron release and absorption. Recent studies demonstrated that the hypoxia-inducible factor (HIF)-1 alpha contributes to the (down-) regulation of hepcidin, which was suggested to be a direct transcriptional mechanism [ 11 ] or mediated by muscle-derived soluble haemojuvelin, which may be increased by the HIF-dependent induction of furin activity [ 12 ]. However, the molecular mechanisms of the hypoxic or anaemic regulation of hepcidin are far from being understood. Several studies demonstrated that the induction of erythropoiesis and not hypoxia or anaemia itself down-regulates hepcidin [ 8 , 11–13 ]. ESA increase the erythropoietic activity, and thus iron must be rapidly mobilized from the stores to satisfy the needs of the bone marrow. The reported decrease in circulating hepcidin levels by ESA treatment [ 6 , 8 , 13 ] may explain the fast iron release. The relationship between hepcidin production and erythropoiesis suggests the presence of a regulator between the erythron and the liver, and several candidates for this role have been proposed, for example the soluble transferrin receptor (sTfR) [ 14,15 ] and the growth differentiation factor (GDF)-15 [ 16 ]. Until recently few investigative tools were available to measure hepcidin in human biological samples. Many studies measured the precursor prohepcidin using a commercially available ELISA. The relevance of these data is controversial because prohepcidin levels do not correlate with urinary and serum hepcidin, nor do they respond to relevant physiological stimuli [ 14 , 17–19 ]. Moreover, in vitro studies do not show a fixed relation between hepatocyte prohepcidin and hepcidin release [ 20 ]. Nevertheless, significant changes in prohepcidin concentration have been reported in ferroportin disease [ 21 ] and in patients with CKD [ 22–24 ]. Ganz and Nemeth et al . have developed a dot blot assay using non-commercially available antibodies for the semi-quantitative measurement of hepcidin in urine [ 25 ]. More recently, new assays have been developed exploiting novel technologies such as surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) and liquid chromatography tandem mass spectrometry (LC/MS-MS) [ 26–30 ]. With the use of internal standards these mass spectrometry techniques enable reliable quantitative hepcidin analysis in urine and serum (our unpublished data). To date, most studies have reported urinary hepcidin concentration. Although these urinary data nicely fit in the concept of hepcidin as the iron regulatory peptide [ 9,10 ] and concentrations of urine and serum highly correlate [ 27 ], we caution against the use of urine hepcidin measurements since urinary secretion of hepcidin will depend on glomerular filtration and tubular reabsorption. Furthermore hepcidin mRNA has been detected in the kidney, which suggests the potential for local production and release into the urine [ 31 ]. Thus, serum and urine levels can technically reliably be measured using various MS-equipment that is more and more exploited in clinical laboratories. However, before the measurement of hepcidin in urine and serum can be introduced into routine clinical practice costs should be substantially reduced, i.e. by further automation of the (pre)analytical procedures and subsequent data-analysis. Also, we should define what body fluid and hepcidin isoform is best suited for diagnostic and therapeutic purposes. Finally, interlaboratory reproducibility should be good. This latter point is currently addressed by our studies that compare the accuracy and precision of the available methods for urine and serum hepcidin analysis. Hepcidin production in patients with CKD will depend on iron status, inflammation, anaemia, hypoxia and EPO (endogenous or exogenous as ESA) (Figure 1 ). Measurements of urinary hepcidin are impossible in many anuric patients with CKD. Due to these limitations in hepcidin measurement, initial studies evaluated serum prohepcidin levels in patients with CKD. Prohepcidin levels were increased and correlated negatively with the glomerular filtration rate [ 22–24 ]. However, since the relevance of prohepcidin is controversial (see above), we seriously doubt if the measurement of prohepcidin contributes to our understanding of the pathophysiology of hepcidin in CKD. Putative regulation of hepcidin in chronic kidney disease. The combined effect of the various pathways determines hepcidin levels. More recently, Tomosugi et al . (semi-)quantified hepcidin-25 in serum using a SELDI-TOF MS-based technology [ 28 ]. These authors observed increased levels of hepcidin-25 in patients on haemodialysis. Hepcidin-25 levels correlated with serum ferritin and Il-6, markers of inflammation. However, hepcidin levels were also higher in patients with normal ferritin levels, suggesting accumulation of hepcidin in renal insufficiency. Hepcidin-25 was removed during dialyses in some but not all patients. The cause of this variability remains unclear, but might be attributed to differences in the membrane of the artificial kidney, residual renal function or induction of hepcidin by the haemodialysis procedure itself. The data of Tomosugi et al . indicated that hepcidin-25 levels were approximately two- to threefold higher in the patients than in controls. These figures must be compared with the 20- to 30-fold increase in dialysis patients of serum β2-microglobulin, a low-molecular-weight protein the excretion of which is almost completely governed by glomerular filtration. Thus, glomerular filtration has a limited influence on hepcidin levels that theoretically might be explained by the existence of a thus far unknown circulating hepcidin carrier that could serve as a mechanism to slow down the renal clearance of hepcidin. Alternatively, increased circulating levels of hepcidin may decrease hepatic hepcidin production through a feedback mechanism. Anaemia is common in patients with renal insufficiency. EPO deficiency is by far the major cause, although shortened erythrocyte survival due to haemolysis, bleeding and oxidative stress may contribute. Most patients with CKD and anaemia can be effectively treated with ESA. However, ∼10% of the patients are hypo- or non-responsive to ESA. Several cohort studies have reported an association between higher doses of ESA and mortality [ 32,33 ]. The important role of ESA resistance is stressed by the results of randomized controlled trials that reported an increased mortality or morbidity in patients who were targeted to high haemoglobin levels [ 34–36 ]. Analysis of the data indicates that these results were driven by the increased mortality rate in those patients who were targeted to a high haemoglobin level, received high doses of ESA, but did not reach the target ([ 34 ], Szczech L, American Society Nephrology meeting 2007, unpublished results). Identification of these ESA-resistant patients is thus pivotal in the proper management of anaemia in patients with CKD. Iron deficiency contributes to ESA resistance. Absolute iron deficiency can be recognized by low ferritin levels and low transferrin saturation (TSAT) and can be easily treated by iron supplements. Most patients with insufficient response to ESA, however, have normal or elevated ferritin levels and TSAT, and are therefore considered to be functionally iron deficient. In functional iron deficiency, as often occurs in (low grade) inflammatory diseases such as CKD, hepcidin levels will be elevated and the release of iron from the reticulo-endothelial cells is inadequate for erythropoiesis. However, it is evident from recent studies that some of these patients respond to intravenously (i.v.) administered iron [ 37 ]. The currently used parameters of iron metabolism, such as TSAT and ferritin, did not predict this response [ 37 ]. Based on the role of hepcidin in iron metabolism and anaemia we propose that measurement of serum hepcidin levels may aid in predicting and monitoring the response to i.v. iron and ESA therapy. Obviously, various studies have shown a good correlation between serum hepcidin and ferritin: serum levels of both parameters are elevated in (low grade) inflammatory diseases such as CKD and decreased in iron deficiency. Thus, controlled studies must evaluate the superiority of hepcidin over ferritin in guiding anaemia treatment in patients with CKD. We envisage that hepcidin has advantages: (1) it directly reflects iron availability and needs for erythropoiesis and (2) it integrates the input from inflammatory and erythropoietic pathways and better reflects the status of iron homeostasis than single parameters such as TSAT, sTfr and CRP [ 14 ]. Recently Kato et al . [ 38 ] evaluated hepcidin as predictor of ESA responsiveness in a small number of dialyses patients. Neither prohepcidin nor hepcidin (measured with the semiquantitative SELDI-TOF method for Hepcidin-25) was valid for this purpose. However, as both assays have their shortcomings, these studies need to be confirmed in larger studies exploiting clinically validated and more quantitative methods. Still, it is unlikely that the measurement of a single value of serum hepcidin before the start of anaemia treatment will be sufficient to predict the responsiveness to i.v. iron and/or ESA with high accuracy. Since hepcidin levels are rapidly regulated, the initial changes in serum hepcidin after the start of i.v. iron or ESA may provide a better parameter to assess the long-term response. In future, hepcidin might also become a target of therapy, since lowering hepcidin may aid in improving the gastrointestinal uptake of iron and its release from macrophages, thus limiting the need for i.v. iron, overcoming functional iron deficiency and improving ESA resistance. In sum, the recent development of reproducible, quantitative assays for serum and urinary hepcidin will enable studies that will define the role of hepcidin, in comparison with more conventional parameters, such as serum ferritin and TSAT in predicting and monitoring the response to ESA and iron treatment in patients with chronic kidney disease. Obviously, if hepcidin proves to be a useful marker of responses to ESA and iron treatment, then cheaper, high-throughput methods, which can be reliably transferred to, and reproduced in, laboratories throughout the world, will be needed to facilitate the routine measurements that clinicians need. Conflict of interest statement . None declared.