Anemia in lymphoma: interleukin-6, hepcidin and erythropoietin

The identification [1] of hepcidin as a key regulator of iron metabolism has opened up new vistas for the investigation of poorly understood anemias, termed the “anemias of chronic disease.” We now understand these conditions to be more accurately described as “anemias of inflammation.” Circumstantial evidence suggests the following scenario. The acute phase cytokine cascade, responding to inflammatory stimuli, up-regulates hepcidin expression in the liver. Via its degradation of ferroportin, the plasma membrane exit channel for iron, hepcidin prevents egress of iron from enterocytes and macrophages of the reticuloendothelial system (RES). Thus, recycled iron from destroyed red blood cells in the RES, and ingested iron internalized into enterocytes, is not released into serum and is, thus, not available for erythropoiesis. Hypoferremia and anemia is the result. Cytokines of the interleukin-6 (IL-6) and bone morphogenetic protein (BMP) families are the important stimulators of hepcidin expression, working through well-defined signal transducer and activator of transcription 3 (STAT-3) binding sites and BMP-responsive elements in the hepcidin promoter.

Since the clinical characteristics of the anemias of chronic inflammation are similar to anemia found in cancer, and since several tumor types are associated with increased cytokine production, especially that of IL-6, hepcidin has been addressed as an explanation for the associated anemia in these conditions. Although the role of hepcidin in anemia of solid tumors is under study [2,3], investigations in hematologic malignancies are especially noteworthy. Indeed, studies in patients with multiple myeloma (MM) [4,5], Hodgkin lymphoma (HL) [6], acute leukemia (AL) [7] and Waldenström macroglobulinemia (WM) [8] demonstrate elevated serum hepcidin levels that inversely correlate with hemoglobin levels. In HL, AL and WM, there were also strong correlations between hepcidin and IL-6 levels, while in the myeloma model, probable up-regulation of serum BMPs (especially BMP-2) was at least as important as IL-6 in mediating enhanced hepcidin expression [5]. The site of hepcidin production is thought to be hepatic in these patients, but hepcidin RNA and protein expression has also been detected in primary WM lymphoma cells [8]. These previous studies in related hematopoietic neoplasms suggest a consistent story incriminating hepcidin, but the report by Tisi et al. in the current issue [9] reminds us that the story is certainly more complicated.

In Tisi's current study of 53 patients with diffuse large B-cell lymphoma (DLBCL), there was a significantly higher concentration of both serum hepcidin and IL-6 versus controls, while erythropoietin production was inadequate for the corresponding degree of anemia. Most importantly, however, amongst these three variables, only IL-6 levels predicted the presence of anemia in a multivariate analysis. IL-6 serum concentrations also correlated with hepcidin levels, albeit modestly (p = 0.03). However, the statistically significant association between anemia (hemoglobin level) and IL-6, but not with hepcidin, suggests that other hepcidin-independent mechanisms are involved which could be induced by IL-6. An alternative explanation is that the correlation between IL-6 and hemoglobin level is an epiphenomenon whereby IL-6 levels are simply markers for some other critical pathway. However, as described by Tisi et al. [9], there is ample evidence for the notion that up-regulated expression of IL-6 itself can cause anemia, namely evidence from IL-6-transgenic mice [10] and therapeutic use of anti-IL-6 antibody in a few patients with Castleman disease [11].

A close inspection of this group's previous study in HL [6] and contrasts with the current DLBCL and previous MM studies are revealing. The frequency and degree of anemia in the HL group was comparable to the current DLBCL group, and the mean and median IL-6 and serum iron levels were also similar. It is true that the serum hepcidin levels were somewhat higher (mean 7.9 nM) in the patients with HL compared to those with DLBCL (6.4 nM), but, most importantly, non-anemic patients with either disease had significantly higher levels of hepcidin than controls. In the HL cohort there was a statistically significant inverse correlation between hepcidin and hemoglobin levels but only in the anemic patients. If the analysis had been performed on the entire group of patients (anemic and non-anemic) it is likely that the correlation would not have been significant, similar to what was found in the patients with DLBCL. Comparisons to the MM studies [4,5] are not easy, because the hepcidin assays used are different. Nevertheless, the data suggest that patients with MM as a whole have a more significant anemia and greater up-regulated serum hepcidin expression (approximately 5-fold over controls [5] compared to 2.2-fold increases in patients with DLBCL [9]). Collectively, these data suggest subtle differences between the patients with MM/WD and the patients with HD/DLBCL. In the former group, a greater stimulation of hepcidin expression appears to be the primary cause of a more severe anemia. This may be due to greater up-regulation of IL-6 production or the involvement of additional cytokines. For example, BMP up-regulation may be specific to the MM model, and BMP-2 markedly synergizes with IL-6 for enhanced hepcidin expression [5]. In the group of patients with lymphoma, however, the mechanism of anemia is more complex. Up-regulated hepcidin expression can contribute to the genesis of anemia but it may not be sufficient. Earlier work suggested that restriction of renal erythropoietin production [12] or inhibition of the erythron's response to erythropoietin [13] could play a role in the anemia of chronic inflammatory disease. Although IL-6 itself has not been indicted in these alterations, other inflammatory cytokines have been, and their heightened levels probably correlate with IL-6 levels, possibly explaining the strong relationship between IL-6 and hemoglobin levels in the lymphoma studies.

The elucidation of anemia etiology in these diseases as well as other malignancies is not simply an academic issue. The use of erythropoietin-stimulating agents (ESAs) for the anemia of malignancies exploded during the 1990s, increasing approximately 10-fold [14], and there is no doubt that ESAs induce a modest rise in hemoglobin level in many patients and a decrease in transfusion requirement. Given the continual stress on the blood supply in blood banks, this has some public health benefit. However, more recent data, demonstrating increases in venous thromboembolism and shortened survival in some patients with cancer, have put a brake on ESA use, especially for hematologic malignancies. In addition, hepcidin antagonists have been identified in the laboratory [15], and clinical trials with hepcidin-targeted agents are forthcoming. The current study by Tisi et al. [9], as well as their prior publication [6], indicate that the tumor type to be tested in clinical trials may be a critical variable. In addition, the complex, multifactorial nature of anemia in some of these patients with lymphoma, involving heightened hepcidin levels, insufficient erythropoietin production and, possibly, additional IL-6-mediated pathways, may dictate an optimal management whereby therapeutics are used in combination.

Potential conflict of interest:

A disclosure form provided by the author is available with the full text of this article at www.informahealthcare.com/lal.