TB or not TB? Soft pity opens the iron gates

In this issue of Blood, Cercamondi et al provide valuable insights into the dynamics of iron handling during the resolution of an inflammatory anemia, by prospectively following patients with tuberculosis (TB) after treatment initiation.¹ 

Anemia of inflammation (AI) is frequently observed during chronic inflammatory and infectious conditions, including Mycobacterium tuberculosis (Mtb) infection.² AI pathogenesis is complex and varies depending on the cause and type of inflammation.³ Mechanisms may involve cytokine-mediated reduction in erythropoietin (Epo) production and responsiveness, impaired maturation of erythroid precursors, and accelerated red cell clearance.³ However, a major contributor is often functional iron deficiency linked to inflammatory upregulation of the iron regulatory hormone hepcidin, most prominently by interleukin-6 (IL-6).³ By inhibiting iron transport via the only known iron exporter, ferroportin, hepcidin blocks dietary iron absorption through duodenal enterocytes and impairs macrophage iron recycling. Chronically elevated hepcidin therefore leads to a deficit in serum iron supply for erythropoiesis: iron may be present in the body but unavailable for utilization because of sequestration. Importantly, competing signals may simultaneously contribute to hepcidin regulation. For example, during erythroid iron demand, Epo stimulates synthesis of the hormone erythroferrone (Erfe) in erythroblasts; Erfe suppresses hepcidin synthesis, facilitating iron release into circulation through ferroportin.⁴ 

How these, and other, stimulatory and suppressive signals are balanced to dictate hepcidin expression and consequent iron handling during complex inflammatory anemias in humans is not well understood. The study by Cercamondi et al provides an in-depth approach to addressing this issue in the context of Mtb-associated inflammatory anemia. The authors carried out a detailed longitudinal characterization of changes in biomarkers related to iron homeostasis and inflammation that occurred following treatment initiation in a cohort of Tanzanian patients (N = 18) with Mtb infection. Importantly, they also investigated whether biomarker changes associated with both intestinal iron absorption and erythrocyte iron utilization through tracking orally and IV-administered stable iron isotope tracers.

Before treatment, most patients were anemic, inflamed, and hyperferritinemic, with elevated serum IL-6 concentrations that associated with raised hepcidin. Accordingly, patients were hypoferremic, characteristic of AI. Epo and Erfe concentrations were also relatively raised, presumably reflecting hypoxia and a degree of erythroid drive, respectively. Whether Epo was “appropriately” raised relative to the extent of anemia, or whether there was a degree of inflammatory suppression of Epo production/responsiveness in this context is not clear. Strikingly, prior to treatment, oral iron absorption was almost completely abrogated, and inversely associated with hepcidin concentration (see figure, panel A).

The authors then examined iron/inflammatory biomarker kinetics over the 6-month treatment course. Notably, a marked decline in acute inflammatory markers with concomitant reduction in hepcidin was observed by 2 weeks posttreatment initiation; given the rapidity of these changes, understanding biomarker dynamics within these first 2 weeks would be useful in the future. During the intensive treatment phase, dilution of IV was administered. ⁵⁴Fe tracer suggested release of stored iron for enhanced erythropoiesis, evidenced by increased reticulocytosis and steady hemoglobin recovery. The authors described slower declines in Epo and Erfe concentrations, although the broad spread of data within this relatively small cohort suggests some caution should be used in interpreting precise kinetics for these parameters. Nevertheless, hepcidin’s association with IL-6 weakened following treatment, whereas there was stronger evidence for its inverse association with Erfe during continuation treatment, consistent with a role for Erfe in accelerating anemia resolution via hepcidin suppression (see figure, panel B). Oral iron absorption increased following treatment, reaching levels typical of adequate absorption by treatment completion.

This study makes several salient points. It provides an unprecedented level of characterization of iron homeostasis during a human inflammatory anemia, being the first to quantify Erfe longitudinally in concert with hepcidin, other iron/inflammatory biomarkers, and iron absorption/utilization in such a context. As such, the authors could examine changes in the balance of key signals affecting hepcidin that occurred following Mtb treatment initiation and link this to iron handling. The data are consistent with a central role for inflammatory-driven hepcidin upregulation during active, untreated Mtb driving near-complete blockade of oral iron absorption. Thus, attempts to overcome iron restriction and treat AI through oral iron during active Mtb or early during treatment would very likely be futile. In contrasting, treating the cause of the inflammation effectively remobilized sequestered iron, correcting the anemia, prior to ultimate restoration of iron absorption capability. Resolving Mtb-associated AI through antiinfective treatment without iron interventions has been described previously,^2,5 although without the finer resolution analysis of iron homeostasis provided by Cercamondi et al.

A similar, albeit milder effect of reduced hepcidin and increased iron absorption accompanying treatment of an infection was previously described by the same authors, who found that Beninese women with asymptomatic malaria absorbed oral iron more efficiently after anti–malaria treatment.⁶ These studies address an important concept for which evidence continues to accumulate: namely, that targeting the causes of chronic infection or inflammation may play a key role in alleviating associated iron deficiency and anemia. Appraising this could be especially relevant for iron intervention–based anemia control programs in populations where even mild inflammation and modestly raised hepcidin are prevalent.⁷ Notably, a recent study, using a Mendelian randomization approach provides evidence that malaria itself causes iron deficiency in African children, strongly suggesting that anti–malarial interventions could help restore iron status.⁸ However, the mechanistic causes of anemia may differ markedly depending on the underlying infection and type of inflammation. In TB, anemia is frequently related to inflammation.² In malaria, anemia is multifactorial, and the contribution of increased hepcidin likely depends on the phase and severity of infection.⁹ In patients with severe COVID-19, inflammation and increased hepcidin can certainly occur, but the causes of COVID-19–associated anemia are likely more multifactorial, including infection of erythroid precursors.^10,11 Therefore, the generality of the concept of targeting the cause of inflammation to allow resolution of infection-associated anemia, without supplemental iron, remains to be established.

Cercamondi et al have elegantly demonstrated that treating a particular infection can be sufficient to suppress inflammation-induced hepcidin, softly open the iron gates, and restore hemoglobin. However, whether this is the case for other types of inflammatory anemia, and not only TB, is the question.