Anemia is a common complication in patients with chronic kidney disease (CKD) (1), which is closely associated with an increased risk of CKD progression and mortality. In the last two decades, erythropoiesis-stimulating agents (ESAs) and iron therapy have always been the cornerstone of renal anemia treatment (2). Although ESAs can effectively correct anemia in most patients, there have been concerns about its potential side effects such as increased risks of tumor, cardiovascular events, and cerebrovascular events (3). Therefore, iron supplementation is used to correct iron deficiency and, meanwhile, to minimize the dosage of ESAs. Even for patients with functional iron deficiency, which means that their stored iron is at a relatively high level, intravenous iron therapy has still been performed in the patients undergoing hemodialysis as long as their ferritin level is lower than 500 or even 800 ng/ml (4, 5). However, this kind of practice is liable to put patients at higher risk for infection, allergy, and even iron overload. Hypoxia-inducible factor-prolyl hydroxylase domain inhibitors (HIF-PHIs) are a new class of agents that can inhibit the degradation of the hypoxia-inducible factor (HIF) to promote erythropoiesis by activating the body’s natural physiological response to hypoxia. It stabilizes and promotes the gene expression of the HIF-1 and HIF-2 by inhibiting the activity of prolyl hydroxylase (PH) enzymes (6). These gene actions include increasing the endogenous erythropoietin (EPO) hormone production and regulating iron homeostasis, and the latter ability seems to be their distinct advantage that is not possessed by ESAs. Several HIF-PHIs have been developed in late-stage global clinical development, and some of them have been approved for clinical application (7). Multiple clinical studies have shown that HIF-PHIs can increase hemoglobin (Hb) levels comparable to the ESAs (8, 9), but its effect on iron metabolism has not been definitively elucidated.

Iron transport disorder is a common cause of refractory anemia in CKD patients (10). CKD patients are subject to be in a chronic inflammatory state, which promotes the abnormal increase of hepcidin levels, resulting in abnormal iron storage and the disorder of the transport of iron. In hyperinflammatory states, this condition even becomes a leading factor in refractory renal anemia. Several large randomized controlled trials (RCTs) have found that HIF-PHIs had a markedly significant effect on iron regulation in the treatment of anemia, mainly manifested as decreased hepcidin, transferrin saturation (TSAT) and ferritin (11), decreased or unchanged serum iron (9), and increased transferrin and total iron-binding capacity (TIBC). These results suggested that HIF-PHIs can improve iron transport and iron utilization in patients. However, it is generally accepted that NDD-CKD populations have a lower risk of chronic inflammatory state compared with dialysis-dependent populations. Some results from several latest RCTs have demonstrated that HIF-PHIs could improve iron metabolism in NDD-CKD populations (12–14). However, the lack of head-to-head comparisons makes it impossible to generalize any conclusions regarding the differences of iron-modulating efficacy between HIF-PHIs and ESAs or among different types of HIF-PHIs. Thus, this study is done to explore whether HIF-PHIs still have therapeutic advantages in regulating iron metabolism in a lower inflammatory state, evaluate the differences in iron-regulating efficacy between different HIF-PHIs, and finally develop evidence-based strategy for the treatment of CKD anemia.

The study protocol is available on the International Prospective Register of Systematic Reviews (PROSPERO) with a registration number of CRD42021242777. Three investigators (Jie Xing, Xiaodong Zhu, and Xiaotong Xie) independently performed study selection and quality assessment according to the study protocol. Any disagreements were resolved by consultation with another investigator (Lina Wang).

This study is the largest network meta-analysis to compare the effects of five HIF-PHIs, two ESAs, and placebo on correcting anemia and iron dysregulation. The ability of all five HIF-PHIs included in this studyto correct anemia is not inferior to ESAs. In addition, desidustat exhibited the strongest Hb-increasing ability. When analyzing the regulatory effects of different kinds of drugs on iron metabolism, this study took five HIF-PHIs as one group and two ESAs as the other group, which made the conclusion more extrapolative. The results of this study showed that, compared with placebo and ESAs, HIF-PHIs could significantly decrease hepcidin, TSAT, and ferritin and increase transferrin and TIBC but did not alter serum iron. This network meta-analysis also observed heterogeneity in the ability of different HIF-PHIs to decrease hepcidin. After calculating the effectiveness of each drug to decrease hepcidin, this study found that daprodustat had the highest hepcidin-lowering efficacy.

ESAs are currently the most widely used drugs for the treatment of renal anemia. Even without considering their potential cardiovascular events and cerebrovascular event risks, some patients who have been in a state of inflammation for a long time would not respond to ESAs, known as ESA hyporesponse (26), making their ability to correct anemia non-ideal. Oxygen-sensing and adaptation mechanisms are among the most important mechanisms of life. Three decades ago, it was discovered that the HIF is a key factor for the body to increase EPO production and correct anemia during hypoxia (27). When oxygen levels drop, HIF-α subunits accumulate and dimerize with the HIF-β to form functional transcription factors HIF-1 and HIF-2, which can directly regulate the expression of the EPO gene and other genes involved in encoding iron absorption–related proteins, thereby correcting anemia. In the presence of sufficient oxygen, the HIF is hydroxylated by HIF-PH enzymes and degraded by ubiquitination (28). HIF-PHIs are a new kind of anemia treatment agent that inhibits the degradation of the HIF by simulating the natural physiological response of the body to hypoxia. A previous network meta-analysis (2,768 patients) (29) reported that HIF-PHIs have an anemia treatment effect similar to ESAs and safer to some extent. However, the accuracy and clinical value of the results of that study were limited because of the small sample size and the lack of direct head-to-head comparisons between different HIF-PHIs. This network meta-analysis enrolled four newly published large studies (these four studies included 1,783 NDD-CKD renal anemia patients), allowing for a greater accuracy of the results and a further extrapolation of the conclusions. In this study, the result showed that the five HIF-PHIs included were non-inferior to ESAs in their ability to correct anemia, and desidustat even showed a stronger ability to increase Hb levels than ESAs.

Iron transport disorder is very common in CKD patients and an important cause of refractory renal anemia, which can be manifested as decreased serum iron and transferrin (30) and increased ferritin after adequate iron supplementation but still low levels of TSAT (31).The iron dysregulation in CKD patients is associated with abnormally increased hepcidin under the inflammatory state (32). Hepcidin is encoded by the HAMP gene, which is located on chromosome 19 q13, and is mainly produced and secreted by hepatocytes. The larger precursor protein produced by hepatocytes undergoes continuous cleavages to finally form a 25-amino acid peptide hormone, which is the main bioactive and predominant form of hepcidin. Hepcidin can modulate iron metabolism by a negative feedback mechanism and is a key hormone for the body to maintain iron homeostasis (33). The decrease of the hepcidin level promotes the release of more iron into the blood from hepatocytes and macrophages, as well as increased intestinal iron uptake, ultimately increasing the iron level in the circulation (30). Can HIF-PHIs ameliorate iron metabolism? Previous meta-analysis has initially observed that HIF-PHIs can decrease hepcidin (34), but there is the lack of exploration of potential differences between different HIF-PHIs. This network meta-analysis found that HIF-PHIs can correct anemia without changing serum iron levels because HIF-PHIs could decrease hepcidin, thereby promoting the transport and utilization of iron. In addition, HIF-PHIs can directly induce hepatic and renal EPO expression and increase the transcription of several iron transport and metabolism genes and finally promote erythropoiesis by improving iron transport (35). This study found that HIF-PHIs had different regulatory effects on iron metabolism, among which daprodustat had the highest hepcidin-lowering efficacy, and enarodustat was second only to it. HIF-PHIs have heterogeneous effects on iron metabolism, possibly because of their different pharmacokinetic profiles and potential drug–drug interactions (7).

The regulatory effects of HIF-PHIs on iron metabolism were also presented in the changes of other iron-related indicators. Iron dysregulation can be divided into absolute and relative iron deficiency according to the level of stored iron (25). Ferritin is an important indicator of the iron storage status of the body. Functional iron deficiency, characterized by elevated ferritin levels that are not available for erythropoiesis, is one cause of ESA hyporesponse in patients with renal anemia (36). This network meta-analysis found that HIF-PHIs decreased ferritin compared to ESAs. However, this did not mean that HIF-PHIs aggravated iron deficiency. Rather, because it promoted the utilization of stored iron (37), the reduction of ferritin by HIF-PHIs was a performance of improved iron transport and bioavailability.

The main function of transferrin is to bind and transport iron in the circulation. TIBC is determined by the level of transferrin, and their increases both imply improved iron transport. TSAT is the ratio of serum iron to TIBC, generally considered as an important biochemical marker of the iron level in the circulation and reflecting iron availability for erythropoiesis. Thus, a decrease in TSAT represents iron deficiency (31). Interestingly, unlike previous studies, this network meta-analysis found that HIF-PHIs could significantly increase transferrin and TIBC and decrease TSAT, without changing serum iron levels, which was a good signal (38). HIF-PHIs increased iron utilization in the bone marrow and failed to increase serum iron levels, but, at the same time, increased transferrin led to elevated TIBC. Therefore, as the ratio of serum iron to TIBC, the reduction of TSAT in this case did not mean functional iron deficiency, but it was a sign of improved iron transport. The said dynamic perspective for analyzing the causation of TSAT changes was obviously different from the traditional static perspective for evaluating iron metabolism. On the contrary, ESAs need to be combined with iron therapy to increase serum iron levels and lead to elevated TSAT, and the elevation of the TSAT in this context did not imply an improvement in iron metabolism. Therefore, ESAs could not improve iron transport but increase the risk of iron overload due to the need for iron supplementation. At the same time, iron supplementation feedback increased the production of hepcidin by hepatocytes, which easily led to the vicious circle of iron dysregulation. In conclusion, in the application of HIF-PHIs, due to its multifaceted role in ameliorating iron transport, the evaluation of iron-related indicators should break through the original static perspective, and it is necessary to dynamically observe and evaluate iron metabolism and transport in patients.

This network meta-analysis had several limitations: firstly, some of the included trials did not use the double-blind method, but, considering that the included outcomes of this study have a unified standard and are not easy to be changed by subjective factors, non-double-blind trials are not excluded. Secondly, the follow-up time between the included trials was quite different; thus, this study could not accurately observe the iron metabolism of patients in different administration stages. However, considering that most HIF-PHIs have not yet been approved for clinical use, this study still included some phase II trials with shorter dosing time. Finally, there were differences in the CKD status of the subjects in the included trials, which might make their iron metabolism status different. However, after calculation, it was found that this difference did not meet the statistical criteria for heterogeneity; thus, this study did not adjust the inclusion and exclusion criteria.

In conclusion, the results of this study suggested that HIF-PHIs should be the first choice in the treatment of NDD-CKD patients with functional iron deficiency. In addition, the potential heterogeneity of HIF-PHIs to correct iron metabolism should be fully considered when formulating a treatment plan.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

XLZ and LW designed this study. JX, XDZ and XX searched the literature. JY analyzed the data and wrote the manuscript. All authors contributed to the article and approved the submitted version.