5. Iron depletion therapy
As a consequence of the abovementioned pathogenic scenario linking iron, inflammation and infections, there is the need to find a possible therapeutic strategy to prevent CRS and onset of fibrosis occurring particularly in patients with COVID-19. The progress in understanding the critical role of pro-inflammatory cytokines in the pathogenesis of other hyperferritinemic syndromes such as MAS and AOSD has led to pilot use of anti-cytokine agents, resulting in an increasing number of successful case reports in patients who were unresponsive to conventional treatments (85). The inhibition of IL-1 (with the use of anakinra and canakinumab) and IL-6 (mainly with tocilizumab) showed a strong efficacy compared to placebo in several cohorts and randomized controlled trials in MAS and AOSD. In a post hoc analysis of data from MEASURE, a randomised, multicentre, double-blind, 24-week, phase 3B trial of tocilizumab in RA, authors depicted a rapid decrease of ferritin, hepcidin and haptoglobin following tocilizumab administration. This is consistent with the idea that IL-6 signalling is a common stimulus to production of these molecules (86). An indirect confirmation of the greater relevance of the IL-6 axis on ferritin levels derives from a recent systematic literature review, performed on patients with MAS while being treated with IL-1 and IL-6 blocking agents. In this review, patients who developed MAS while treated with canakinumab trended towards lower ferritin at MAS onset than the historical cohort. In comparison, patients who developed MAS while treated with tocilizumab were less likely febrile and had notably lower ferritin levels (87). The anti-IL-6 effect on ferritin could explain part of the emerging successful reports on tocilizumab treatment in SARS-CoV-2 infection.
Nonetheless, the rapidity of the onset of inflammation in the acute phase of SARS-CoV-2 infection may provoke increased ferritin production to permit adequate storage of iron and to deprive pathogen of iron. If the binding capacity of transferrin in the blood is exceeded, iron may be found in the plasma as non-transferrin bound iron that changes to its redox active form termed labile plasma iron (LPI) (88). LPI correlates with ferritin levels and contributes to the formation of reactive oxygen species (ROS) resulting in tissue damage and subsequent fibrosis (89) (figure 1). Thus, a novel approach to COVID-19 treatment can be represented by iron chelation therapy that can interrupt these steps. Iron chelation represents a pillar in the treatment of iron overload due to a wide spectrum of diseases and multiple chelating agents are currently registered and routinely used in clinical practice. Indeed, deferoxamine (DFO) has a direct effect on ferritin since promotes ferritin degradation in lysosomes by inducing autophagy, while both deferiprone and deferasirox are likely to chelate cytosolic iron and iron which is extracted from ferritin prior to ferritin degradation by proteasomes (90) (Figure 1). Moreover, several studies have been performed on the potential anti-viral effect of iron chelating therapy. Indeed, iron overload can contribute to human immunodeficiency virus (HIV) replication in vitro by increasing reverse transcriptase activity and reducing the viability of infected T cells. Iron chelation by DFO has shown beneficial effects on HIV infection (91) probably through multiple mechanisms such as: 1) restriction of DNA synthesis through the inhibition of ribonucleotide reductase, which requires iron to exert its enzymatic activity, 2) inhibition of T cell proliferation that is essential for HIV replication, 3) direct toxic effect on viral DNA and RNA via oxidative stress and 4) inhibition of NF-kB pathway. These effects may not be universal for all iron chelating agents. In fact, DFO and deferiprone (DFP) can both inhibit T cell proliferation and DNA synthesis, while bleomycin can directly bind to viral DNA with no effect on host T cells (92, 93).
A potential anti-viral effect has also been demonstrated with other pathogens, such as HSV-1 (94) and CMV. More specifically, CMV requires iron in order to induce the increase in size of infected cells, so that increased in vitro levels of free iron have been demonstrated before the occurrence of this phenomenon, which can be effectively limited by iron chelation therapy (95). DFO is also capable to further enhance the therapeutic effect of IFN on hepatitis B virus (HBV) infection (96). Fewer data are available on the effect of iron chelation on other infective agents, though Mateos et al. (97) reported increased levels of free iron in the bronchoalveolar lavage (BAL) fluid of HIV patients with Pneumocystis jiroveci pneumonia compared to controls, suggesting a potential pathogenic role of iron. Similarly, a beneficial effect of DFO treatment was demonstrated in a murine model ofTrypanosoma cruzi infection, independently from the iron metabolism of the host cell (98). However, it should be carefully considered that iron chelators may actually be exploited by pathogens as sources of iron (99), thus a careful analysis of the pharmacodynamic mechanisms of the single chelating agents available is warranted.
One of the main mechanisms through which iron can promote inflammation is mediated by an increased production of free oxygen radicals via Haber-Weiss reaction. As an example, iron is able to increase the in vitro production of IL-6 by endothelial cells following infection with Chlamydia pneumoniae and Influenza A virus, which can be effectively controlled by DFO (100). Interestingly, similar processes, including IL-6 and free oxygen radical production, take place during septic shock. Thus, it is not surprising that iron chelation is effective in decreasing mortality in murine models of septic shock via NO scavenging (101) and inhibition of MAP kinases and NF-kB pathways, eventually leading to reduced production of pro-inflammatory cytokines (102).
One of the most severe complications of diseases leading to iron overload is liver damage, characterised by progressive fibrosis and, eventually, irreversible cirrhosis. In fact, the prevention of liver damage is the main indication for iron chelation in these conditions. Although the reduction of free iron levels and, consequently, of oxygen radicals, is the main mechanism preventing progressive damage, some authors suggested iron chelating agents may exert an independent anti-fibrotic effect. This evidence comes from studies showing reduction of liver fibrosis in the absence of a significant decline in liver iron content (103). Deferasirox (DFX) and DFO seem able to reduce damage and fibrosis in multiple rat models of concavalin A and CCl4-induced liver injury by inhibiting the production of free radicals (104–106), though other studies did not confirm this evidence (107). Anti-fibrotic effects in kidney disease have also been demonstrated in rat and mouse models of renal damage, again via a reduction of oxidative stress, macrophage tissue infiltration and production of pro-fibrotic cytokines such as TGF-β (108, 109). Other authors showed that DFO can provoke a remarkable decrease of IL-6 levels and have a potent anti-fibrotic effect in HCV infection (110).
Whether these phenomena share common aspects with COVID-19 is currently not known. It is, however, reasonable to speculate that iron chelation may influence free radicals and pro-inflammatory cytokines production that are strongly involved in the late phase of COVID-19, eventually leading to acute lung injury and ARDS. It has been shown that mechanical ventilation, often required in COVID-19 patients, may induce lung injury that is known to be associated with the release of inflammatory factors, apoptosis, endothelial dysfunction, and activation of the coagulation system (111, 112). Interestingly, pre-conditioning with DFO showed a lung protective effect against mechanical ventilation through effective reduction of ROS formation in macrophages and mitochondria in a mouse model (113).
Additionally, preliminary data seem to suggest that residual lung damage may be present in a subset of severe COVID-19 patients following the acute phase of the disease (114). If these data were to be confirmed, the anti-fibrotic effect of iron chelating agents may represent an additional mechanism of action deserving careful consideration.
To conclude, the abovementioned considerations lead to the idea that COVID-19 may be part of the hyperferritinemic syndrome spectrum (115). Possible iron acute overload caused by rapid synthesis of ferritin exceeding its iron incorporation rate, and the beneficial effects of iron chelation therapy on the inflammatory status as well as on the fibrogenesis occurring in the lungs suggest that, in appropriate setting of critically ill patients with COVID-19, iron chelation therapy could be considered to improve survival and overall long-term outcome.